{ "complete": true, "report": { "bucket": "vn_TENS_reg_low", "label": "very relaxed and unstrained", "definition": { "id": "vn_TENS_reg_low", "category": "bipolar", "head": "vn_TENS_reg", "pole": "low", "label": "very relaxed and unstrained", "floor": 1.0, "canonical_dimension_name": "Tension" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P02", "general": "Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "inline": [ "Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality." ], "recipe": "P02", "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": 4.0 }, "n": 20, "target_z": 0.16303711049400257, "target_raw": 1.5237490743398667, "ce": 6.008118224143982, "genuineness": 4.375236475467682, "identity": 0.6143898449838161, "wer": 0.03571529927760578, "valid_fraction": 0.95, "identity_delta": 0.035881357640028, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 0.26623535592661, "target_raw": 1.4378593206405639, "ce": 5.919557428359985, "genuineness": 4.371784639358521, "identity": 0.5959419630467891, "wer": 0.04657163742690058, "valid_fraction": 0.9, "identity_delta": 0.017433475703001022, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 0.16303711049400257, "target_raw": 1.5237490743398667, "ce": 6.008118224143982, "genuineness": 4.375236475467682, "identity": 0.6143898449838161, "wer": 0.03571529927760578, "valid_fraction": 0.95, "identity_delta": 0.035881357640028, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 0.1505859348143938, "target_raw": 1.5341119289398193, "ce": 5.899700784683228, "genuineness": 4.353584158420563, "identity": 0.561120331287384, "wer": 0.07160281217750258, "valid_fraction": 0.9, "identity_delta": -0.017388156056404112, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": -0.0022399960811634633, "target_raw": 1.6613057732582093, "ce": 6.133016729354859, "genuineness": 4.4546283960342405, "identity": 0.6355183392763137, "wer": 0.012916666666666667, "valid_fraction": 1, "identity_delta": 0.05700985193252563, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": -0.004687500000000005, "target_raw": 1.6633427798748017, "ce": 6.130832552909851, "genuineness": 4.361115324497223, "identity": 0.5849876791238785, "wer": 0.011822755417956656, "valid_fraction": 1, "identity_delta": 0.006479191780090332, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": -0.01607666882851997, "target_raw": 1.6728217482566834, "ce": 6.162801790237427, "genuineness": 4.418227338790894, "identity": 0.5957716077566146, "wer": 0.010818713450292398, "valid_fraction": 1, "identity_delta": 0.017263120412826537, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.09819336049400258, "target_raw": 1.7411657631397248, "ce": 6.173722958564758, "genuineness": 4.35860493183136, "identity": 0.5949737146496773, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.01646522730588913, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.17094729136205924, "target_raw": 1.801717346906662, "ce": 6.15406482219696, "genuineness": 4.384022951126099, "identity": 0.554144636541605, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.024363850802183153, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.07350771248802578, "target_raw": 1.7206204116344452, "ce": 6.176900005340576, "genuineness": 4.349190950393677, "identity": 0.5235905140638352, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.054917973279953, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 0.26623535592661, "target_raw": 1.4378593206405639, "ce": 5.919557428359985, "genuineness": 4.371784639358521, "identity": 0.5959419630467891, "wer": 0.04657163742690058, "valid_fraction": 0.9, "identity_delta": 0.017433475703001022, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 0.2575195254311866, "target_raw": 1.4451133251190185, "ce": 5.787311935424805, "genuineness": 4.365508580207825, "identity": 0.5776188731193542, "wer": 0.05992367561059511, "valid_fraction": 0.95, "identity_delta": -0.0008896142244338989, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 0.21538085333238607, "target_raw": 1.4801844656467438, "ce": 5.995211386680603, "genuineness": 4.387788605690003, "identity": 0.5623659506440163, "wer": 0.012013888888888888, "valid_fraction": 1, "identity_delta": -0.016142536699771882, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 0.17716064537050194, "target_raw": 1.5119943499565125, "ce": 5.894538044929504, "genuineness": 4.411323690414429, "identity": 0.5856886774301528, "wer": 0.038133169934640526, "valid_fraction": 0.95, "identity_delta": 0.0071801900863647464, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 0.17229001567534616, "target_raw": 1.5160480737686157, "ce": 5.939017009735108, "genuineness": 4.333186936378479, "identity": 0.5682150170207023, "wer": 0.02219298245614035, "valid_fraction": 0.95, "identity_delta": -0.010293470323085785, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 0.14494628357938738, "target_raw": 1.5388056933879852, "ce": 5.954108023643494, "genuineness": 4.326910996437073, "identity": 0.5876930132508278, "wer": 0.05884868421052632, "valid_fraction": 0.95, "identity_delta": 0.009184525907039642, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 0.1435180638884942, "target_raw": 1.5399943709373474, "ce": 6.052795767784119, "genuineness": 4.375236475467682, "identity": 0.6127928882837296, "wer": 0.02175438596491228, "valid_fraction": 0.95, "identity_delta": 0.03428440093994141, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.021728509134785065, "target_raw": 1.641357284784317, "ce": 6.140562295913696, "genuineness": 4.370529389381408, "identity": 0.5640526577830315, "wer": 0.028004385964912278, "valid_fraction": 0.95, "identity_delta": -0.014455829560756684, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 0.014007577563171184, "target_raw": 1.6477832555770875, "ce": 6.073645234107971, "genuineness": 4.311220860481262, "identity": 0.6016749605536461, "wer": 0.011867690058479532, "valid_fraction": 1, "identity_delta": 0.02316647320985794, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.009240719690893167, "target_raw": 1.6517506122589112, "ce": 6.158433270454407, "genuineness": 4.435486483573913, "identity": 0.6211193814873696, "wer": 0.008888888888888889, "valid_fraction": 1, "identity_delta": 0.04261089414358139, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.006347662964015457, "target_raw": 1.6541584432125092, "ce": 6.126662611961365, "genuineness": 4.375864052772522, "identity": 0.5717661246657372, "wer": 0.00625, "valid_fraction": 1, "identity_delta": -0.006742362678050995, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 0.004528798578676946, "target_raw": 1.6556722462177276, "ce": 6.097274708747864, "genuineness": 4.390612769126892, "identity": 0.6174478113651276, "wer": 0.009027777777777777, "valid_fraction": 1, "identity_delta": 0.038939324021339414, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": -0.010314926355665355, "target_raw": 1.6680263698101043, "ce": 6.125471186637879, "genuineness": 4.411323702335357, "identity": 0.6126277491450309, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.03411926180124283, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": -0.046740726852509656, "target_raw": 1.6983428061008454, "ce": 6.151284885406494, "genuineness": 4.37931592464447, "identity": 0.5766946256160737, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.0018138617277145385, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": -0.0734985384573076, "target_raw": 1.7206127762794494, "ce": 6.174715781211853, "genuineness": 4.418541169166565, "identity": 0.6232605934143066, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": 0.044752106070518494, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.09019776660692923, "target_raw": 1.7345111966133118, "ce": 6.190998268127442, "genuineness": 4.375550293922425, "identity": 0.5846535429358483, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.006145055592060089, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.10716550524344913, "target_raw": 1.7486330926418305, "ce": 6.179878520965576, "genuineness": 4.42764139175415, "identity": 0.5708885379135609, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": -0.00761994943022728, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.13822020796188417, "target_raw": 1.7744792759418488, "ce": 6.199536609649658, "genuineness": 4.415089273452759, "identity": 0.5922532707452774, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.013744783401489259, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.14133301172900903, "target_raw": 1.7770699977874755, "ce": 6.158036184310913, "genuineness": 4.389671325683594, "identity": 0.5698220640420913, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.008686423301696777, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.146020511729009, "target_raw": 1.7809713065624238, "ce": 6.183452701568603, "genuineness": 4.3852781534194945, "identity": 0.550812178850174, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.027696308493614197, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.177239974596314, "target_raw": 1.8069546163082122, "ce": 6.18007709980011, "genuineness": 4.303062009811401, "identity": 0.5766758233308792, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0018326640129089355, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.20260007499644783, "target_raw": 1.8280613005161286, "ce": 6.217209053039551, "genuineness": 4.384336686134338, "identity": 0.5771441832184792, "wer": 0.018421052631578946, "valid_fraction": 0.95, "identity_delta": -0.0013643041253089904, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.20322267012563194, "target_raw": 1.8285794734954834, "ce": 6.17511293888092, "genuineness": 4.394692146778107, "identity": 0.6100081011652947, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0314996138215065, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.09302978487649936, "target_raw": 1.7368682265281676, "ce": 6.185041236877441, "genuineness": 4.403792452812195, "identity": 0.5238014560192823, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.05470703132450581, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.11367799281088291, "target_raw": 1.7540533006191255, "ce": 6.182658457756043, "genuineness": 4.385278177261353, "identity": 0.5785084873437881, "wer": 0.0027777777777777775, "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": -0.06091307888778377, "target_raw": 1.7101381599903107, "ce": 6.186232662200927, "genuineness": 4.328166139125824, "identity": 0.5726406127214432, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.005867874622344971, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -0.061816402221590724, "target_raw": 1.7108899772167205, "ce": 6.23110876083374, "genuineness": 4.403478670120239, "identity": 0.5451408833265304, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.03336760401725769, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.13293456259138223, "target_raw": 1.770080143213272, "ce": 6.112365794181824, "genuineness": 4.204214346408844, "identity": 0.5470367535948754, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03147173374891281, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.2193212821470641, "target_raw": 1.8419780135154724, "ce": 6.243419923782349, "genuineness": 4.401030983924866, "identity": 0.5999285724759101, "wer": 0.011057312252964427, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": -0.0640234448406569, "target_raw": 1.7127268528938293, "ce": 6.1396885681152344, "genuineness": 4.521907558441162, "identity": 0.6108118721842766, "wer": 0.016721249764728024, "valid_fraction": 1, "identity_delta": 0.010883299708366394, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 0.03870116350486073, "target_raw": 1.627231297492981, "ce": 6.107679615020752, "genuineness": 4.533455410003662, "identity": 0.6112429520487785, "wer": 0.013146740368708332, "valid_fraction": 1, "identity_delta": 0.011314379572868348, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 0.0012207175825132788, "target_raw": 1.6584254932403564, "ce": 6.097512950897217, "genuineness": 4.590567426681519, "identity": 0.6102218699455261, "wer": 0.014564276303406738, "valid_fraction": 0.98, "identity_delta": 0.010293297469615936, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.037480445922347456, "ci95_z": [ 0.009824196638509104, 0.06809568034210756 ], "language_deltas": { "en": -0.001416037781251002, "de": 0.09570310172474643, "fr": 0.046289036539140245, "nl": 0.03974606913194333, "es": 0.0070800599971582775 }, "languages_improved": 4, "clear_gain": true, "delta_ce": 0.010166664123535157, "delta_genuineness": -0.057112016677856446, "delta_wer": -0.0014175359346984058, "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": 0.10272460834551764, "ci95_z": [ 0.07494140938320722, 0.12732908945489213 ], "language_deltas": { "en": 0.08935550321527934, "de": 0.10461425814820079, "fr": 0.12363280444318146, "nl": 0.05390624641919176, "es": 0.14211422950173483 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.03200895309448242, "delta_genuineness": 0.0115478515625, "delta_wer": -0.003574509396019693, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.15529783730640717, "ci95_z": [ 0.12916503572787233, 0.18244138646603447 ], "language_deltas": { "en": 0.14628902221590728, "de": 0.19492185530555467, "fr": 0.14799802740435516, "nl": 0.18115234419760104, "es": 0.10612793740861777 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.10373135566711426, "delta_genuineness": 0.12087657451629638, "delta_wer": 0.0056639375117635985, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": true, "scale": 1.0, "cfg": 5.0, "reason": "Held-out target improves against matched CFG and best manual-only configuration; QA passes" }, "reference_only": true, "updated": 1793046749.2311437, "example": { "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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.", "language": "nl", "seed": 1701, "scale": 1.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "00648afd5157ee757540f5e944ff1b19a03e570d24d5f487caa0fdf0186216b7", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "813d466330409b2108f7c4323682e9c12d6942f2e3b1ad9604831db1f03e05e8", "generation_seconds_batch": 15.988453065045178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7691657543182373, "target_z": -0.13183590616792787, "ce": 6.268836498260498, "genuineness": 4.808597087860107, "identity": 0.5547168254852295, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/00648afd5157ee757540f5e944ff1b19a03e570d24d5f487caa0fdf0186216b7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "00dd09fc7efbf3c47c5453acaa66c4a6ba99fad4c97bbd0589099ab289e84984", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eecb0a62596673de124a129da8f0684166897edc6f727dc37f745d508f5418b5", "generation_seconds_batch": 23.72884730703663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2543911933898926, "target_z": -0.7148439111363709, "ce": 5.692992687225342, "genuineness": 4.344170093536377, "identity": 0.5841670036315918, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/00dd09fc7efbf3c47c5453acaa66c4a6ba99fad4c97bbd0589099ab289e84984.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "00e7862076a00dbc4ebec0c4b5bfa0a6c3d461c634f3adec644f001e5bd61828", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e69cf5bcb8cf4658e58eeb26b06b8a387d19c1ebdd9a4e8f057269350a50108d", "generation_seconds_batch": 24.001254394999705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6347534656524658, "target_z": 0.02966312901909917, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.4515971541404724, "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_TENS_reg_low/00e7862076a00dbc4ebec0c4b5bfa0a6c3d461c634f3adec644f001e5bd61828.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "015078afe7ee793f97835a5a2721a40263b71bc39ebe7fffde7a9f2073a19bfe", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "150123b37bb01c7725a7f2e5a2898439df5383301aacd0fe1a5939c1ce9256dc", "generation_seconds_batch": 24.261832074960694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2205442190170288, "target_z": 0.5273437679040412, "ce": 6.109982967376709, "genuineness": 4.319066047668457, "identity": 0.41913485527038574, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam.", "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_TENS_reg_low/015078afe7ee793f97835a5a2721a40263b71bc39ebe7fffde7a9f2073a19bfe.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very relaxed and unstrained. 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": "2cba9da9ed053457687a213847dec06bb3ec1fe8c264bce1a12113fc12f96f41", "generation_seconds_batch": 17.60408251208719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.568817377090454, "target_z": 0.1088866493718403, "ce": 5.982900142669678, "genuineness": 4.362998008728027, "identity": 0.8320900201797485, "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_TENS_reg_low/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "023934746d8274bbe849493823cb8da50ba249ef655253ae6392f002a4c0e369", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc164960b6880a5e468cdb020aa9670bcf2406c0f575d4b650f649c002cf0af8", "generation_seconds_batch": 26.575422744033858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0222277641296387, "target_z": 0.7656249283838351, "ce": 5.911416053771973, "genuineness": 4.287685871124268, "identity": 0.6724507212638855, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/023934746d8274bbe849493823cb8da50ba249ef655253ae6392f002a4c0e369.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "0284fc489424b02f5b0c269a068b75fc53f0df92f132eae548c809a2ff20f087", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "214cf4d3cdb41a0915005114a80002dd0b3cb0172fd76b4dbcf487e562aee8fd", "generation_seconds_batch": 18.61482088593766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8317493200302124, "target_z": -0.20703130371212364, "ce": 6.038498878479004, "genuineness": 4.250029563903809, "identity": 0.46281367540359497, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0284fc489424b02f5b0c269a068b75fc53f0df92f132eae548c809a2ff20f087.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "029ba550c662c0d7001b26ff882d1c7c2082d384669f91cb90c2eb2047ccd4e3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "30dc5a3d5c18ddda9e1e4c543344b3220c522e7da1fdea580d328eaac4f1f33d", "generation_seconds_batch": 26.032341446029022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.087249517440796, "target_z": 0.6875, "ce": 5.403085231781006, "genuineness": 4.080576419830322, "identity": 0.7554771304130554, "hypothesis": "J'ai trouvé la petite boite. Elle était prêt de la vie fenêtre.", "wer": 0.5263157894736842, "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_TENS_reg_low/029ba550c662c0d7001b26ff882d1c7c2082d384669f91cb90c2eb2047ccd4e3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "02b007d0fdce567aef5351c7cf33b3c6c344767ac0808b74665511d333edca0e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "74ee8f7caee35883a785166a02348bad4b155bd1246f701ecbfeb02895a5da9a", "generation_seconds_batch": 18.61055895802565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6923587322235107, "target_z": -0.03955074320391242, "ce": 6.3641486167907715, "genuineness": 4.055472373962402, "identity": 0.6785908937454224, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/02b007d0fdce567aef5351c7cf33b3c6c344767ac0808b74665511d333edca0e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "02d80ec8cf33d900e00fe049a9e16621ef33c04d3b063183dcd8231a07ffdff2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b3483215009630f9132a2b1087a623ac879ec235455a8bb1aab6fe96a49d5d5", "generation_seconds_batch": 23.27146893402096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.230297565460205, "target_z": 0.5156249283838351, "ce": 6.252951145172119, "genuineness": 4.444586753845215, "identity": 0.45669519901275635, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/02d80ec8cf33d900e00fe049a9e16621ef33c04d3b063183dcd8231a07ffdff2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0319b812043cc33e0dc838bc12fe53114775bf0010da20742291395297dc597e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a9530b1b715f25e623f67709bf579afb70c26ecb0a239eb214bca422ce2e8f53", "generation_seconds_batch": 18.125881711021066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7386868000030518, "target_z": -0.09521483703598455, "ce": 6.181467056274414, "genuineness": 4.7081804275512695, "identity": 0.6477959156036377, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0319b812043cc33e0dc838bc12fe53114775bf0010da20742291395297dc597e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "033130c0be14b9effeddea8439b7362123fda8dddd54d68bd6221bace7ca25dd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "dfa1f42fee46e8621a80f4893d5a5832a893ede3b72f39533888234302f113f6", "generation_seconds_batch": 26.772241360973567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1197603940963745, "target_z": 0.6484375358080824, "ce": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.16459326446056366, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/033130c0be14b9effeddea8439b7362123fda8dddd54d68bd6221bace7ca25dd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "033a14a5722d204c70b34fa9c5c08b980498ab60136f050a3814555615d70d2e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8340da4fc2756c39dc99fcb405b4129dc8198932a8777b7ecb30dad2b4057bdd", "generation_seconds_batch": 17.96617010398768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.166611671447754, "target_z": -0.6093750716161649, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.6704105734825134, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/033a14a5722d204c70b34fa9c5c08b980498ab60136f050a3814555615d70d2e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "034bf886444ea00b124f84ca9f65630f8c8481883e26304c8785de62a6989143", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "536c0dbbf2a9227118496e561d9e0415e33e7e377511298390c5b0cc6aaecd6b", "generation_seconds_batch": 24.95824826299213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.155522346496582, "target_z": 0.605468839520206, "ce": 5.71682071685791, "genuineness": 4.632867813110352, "identity": 0.6561450958251953, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/034bf886444ea00b124f84ca9f65630f8c8481883e26304c8785de62a6989143.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0395a40a3b1768d4646ba2f1589d159f5e89b3cd582bd17b168caa7829c2d0b4", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "48c008777ecd94250e70f2c55b6cb2f23af6e848467cedd261378dbc4b6cc292", "generation_seconds_batch": 16.30997991899494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8350003957748413, "target_z": -0.21093753580808242, "ce": 6.0305562019348145, "genuineness": 4.607763767242432, "identity": 0.7391942143440247, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0395a40a3b1768d4646ba2f1589d159f5e89b3cd582bd17b168caa7829c2d0b4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "042fcb5d5503e90e69de777829c6d2665c58af622f0eb5764fdf94f5461eaf67", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbad7462d1306ae62f12c2dc6903e7854b15251290ac0fdb7f24920d294cebe1", "generation_seconds_batch": 17.727184303919785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2400507926940918, "target_z": 0.5039062320959588, "ce": 6.1894097328186035, "genuineness": 4.7081804275512695, "identity": 0.8891419172286987, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/042fcb5d5503e90e69de777829c6d2665c58af622f0eb5764fdf94f5461eaf67.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "04aee4ae4c91998c0d94efb425806a240470958feff57b19cc03d43f4417afb8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "755da3ae8cb1cd5da24e6ad436b84c433279301e82959cd0c2afe09216923c51", "generation_seconds_batch": 16.250440262025222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5090785026550293, "target_z": 0.18066409383207213, "ce": 6.022613525390625, "genuineness": 4.218649387359619, "identity": 0.78328937292099, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/04aee4ae4c91998c0d94efb425806a240470958feff57b19cc03d43f4417afb8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "04b331c362e676597cc65fb1f754886d46c5caeb4401bd82838938e74e16427c", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "f7fed310380ca0a2f11c43a40c1d15aadecc186539a756e15fdc496ec166e1cb", "generation_seconds_batch": 16.30997991899494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7699785232543945, "target_z": -0.13281246419191758, "ce": 6.09409761428833, "genuineness": 4.65797233581543, "identity": 0.7486254572868347, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/04b331c362e676597cc65fb1f754886d46c5caeb4401bd82838938e74e16427c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "05017d9b7d6d8b70cc97f35e2945f065d5f333cfcd6c363d33c8ca7d082c2530", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "695160150e871cc376f2f32bf6769de5794293645e9755f3c261cbd111c36559", "generation_seconds_batch": 26.032341446029022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1327648162841797, "target_z": 0.6328124641919176, "ce": 5.863759994506836, "genuineness": 4.7081804275512695, "identity": 0.6518876552581787, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/05017d9b7d6d8b70cc97f35e2945f065d5f333cfcd6c363d33c8ca7d082c2530.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "05129e036dc1b20b49108a2619e1cb80146dfb7010d95395460d690a937a2b64", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "125c7c2c98af75f02b4e1f912e37ce73fa616bc213b3f3b1735eeff03f4e0469", "generation_seconds_batch": 17.48881007207092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.082083225250244, "target_z": -0.5078124641919176, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.4151301085948944, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/05129e036dc1b20b49108a2619e1cb80146dfb7010d95395460d690a937a2b64.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "0519d35092bedfee83dd6616b76cd1723b4f73d31107d4debe544b52354cc434", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9416a8757c8044fb8e8170bc268df800151baf16ea0e111e1d78c8ff4910bb55", "generation_seconds_batch": 21.44056558900047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4269884824752808, "target_z": 0.2792968839520206, "ce": 6.1894097328186035, "genuineness": 4.18099308013916, "identity": 0.6150315999984741, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0519d35092bedfee83dd6616b76cd1723b4f73d31107d4debe544b52354cc434.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "0582ecf774787b2e51fd1962148591401f74c6892c8475ff5895eb748f3c1a6a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74c095eca9d74b12f5a9f30e943971efee712db3c5f84160f54a7e8d75a01c41", "generation_seconds_batch": 16.246735175023787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2449274063110352, "target_z": 0.4980468839520206, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.6707000732421875, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0582ecf774787b2e51fd1962148591401f74c6892c8475ff5895eb748f3c1a6a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "058ef0897e4e5bb9651945f91a32b0e4cd07832df79d6b5e79f919efbcd7469a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8a8db38ae109ab0b8b5e2445a8932a904be34980b61ca23497657d935bb6712b", "generation_seconds_batch": 18.0726844179444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6737818717956543, "target_z": -1.2187501432323298, "ce": 6.109982967376709, "genuineness": 4.111956596374512, "identity": 0.25545600056648254, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/058ef0897e4e5bb9651945f91a32b0e4cd07832df79d6b5e79f919efbcd7469a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "05921340d17613517fd9f9be55e1603fc9ce8f50967673a898795f035c4fb933", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a91a3364bdd1a7c4e6e3e44b2f14de65901c03ace13ddff2ec9ce40dd669309", "generation_seconds_batch": 25.33220457204152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9845504760742188, "target_z": -0.39062492838383517, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.6307017803192139, "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_TENS_reg_low/05921340d17613517fd9f9be55e1603fc9ce8f50967673a898795f035c4fb933.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very relaxed and unstrained. 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": "375ecd2c1dd79cfd19c24bc209d8459483c54d1f5a5f718c585482ac7a4830db", "generation_seconds_batch": 18.48532224004157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.960167407989502, "target_z": -0.36132818766414426, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.6870507597923279, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sagt mir, wann wir reden können.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "05cf41dd4b6ad16a4175bdeccbe08dc1119364cb23280f5db9ab6106c0349406", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "dc5ad50c2e2324e7d0a7ea97df5e948211fab8287e2de4a587f8e041f5550c97", "generation_seconds_batch": 18.0726844179444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.342170476913452, "target_z": -0.8203124641919176, "ce": 5.982900142669678, "genuineness": 4.193545341491699, "identity": 0.7203103303909302, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/05cf41dd4b6ad16a4175bdeccbe08dc1119364cb23280f5db9ab6106c0349406.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "05fa478d5d0d60858a5943edf38945011f8548740ce387cec0dc67a80d5c069f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5cb666d8f157f37d39cba0ba28f10e5ad64bbe42b361ffe45caa03c79017926", "generation_seconds_batch": 24.677046103053726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9991804361343384, "target_z": -0.4082031160479794, "ce": 6.213237762451172, "genuineness": 4.482243061065674, "identity": 0.5626437664031982, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/05fa478d5d0d60858a5943edf38945011f8548740ce387cec0dc67a80d5c069f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "06028b323723a9cd2d3a5bbbb9566f0f83fc47e9f1c6d2ec46dc0c40d58bd816", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e7aa8af40fe9d50def7d4a7771101f9538181c074e8f7531ce1027d2c6ca3eb", "generation_seconds_batch": 24.63003994396422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4725037813186646, "target_z": 0.22460934814393818, "ce": 5.927301406860352, "genuineness": 4.344170093536377, "identity": 0.5835800170898438, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/06028b323723a9cd2d3a5bbbb9566f0f83fc47e9f1c6d2ec46dc0c40d58bd816.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "06569469fde8779ea97d3922dfc0dd04613a803be1af3abc26504abc173399cd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25907821cf9bd9d5569f5e6b6e3d827e06e1f02889e5c57b2f9a1d5814481885", "generation_seconds_batch": 24.842032745014876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.577351450920105, "target_z": 0.09863279011994848, "ce": 6.221180438995361, "genuineness": 4.570107460021973, "identity": 0.7960419654846191, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/06569469fde8779ea97d3922dfc0dd04613a803be1af3abc26504abc173399cd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "06ff8bf75a6c4d9ce29c2aec54b5a8a93febb68b8c15e6df163cf3b77b96848b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "006e414c8cd10bbac1264f944517fdc0e0486fe7f1e6036281284f0dbf81f883", "generation_seconds_batch": 18.744288856978528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8967710733413696, "target_z": -0.2851562320959588, "ce": 5.903473377227783, "genuineness": 3.9299514293670654, "identity": 0.24728208780288696, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are", "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_TENS_reg_low/06ff8bf75a6c4d9ce29c2aec54b5a8a93febb68b8c15e6df163cf3b77b96848b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "07168fb1566b20d134fd0c37f8f0a4000dafaaca30acd78ca645ba35d344f89b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3d13f61ca5a97245f0e76af3fa1a299aea8209eb3e9153565490e80cc999b88", "generation_seconds_batch": 23.018153270008042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4026052951812744, "target_z": 0.3085937679040412, "ce": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.3362051248550415, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/07168fb1566b20d134fd0c37f8f0a4000dafaaca30acd78ca645ba35d344f89b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "072387ab25532adc0b218ea25dbba2f664283dfaf5b7d69e87b0677c3507b103", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a26215ac63baf7013be52f8e84e0936d73530e997344b284ce60fc2672652be", "generation_seconds_batch": 24.85925367800519, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5992963314056396, "target_z": 0.07226558023989696, "ce": 6.157639026641846, "genuineness": 4.356721878051758, "identity": 0.3502618670463562, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/072387ab25532adc0b218ea25dbba2f664283dfaf5b7d69e87b0677c3507b103.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "0812f93876d8c6844dc00a32e435f1e6266bac45b3a673cdc5b43e9b9cfd26e1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1fcc5d27a8010cac285e8d9e1e5380248616eb15507884dbfd995f26c00e187f", "generation_seconds_batch": 24.186294093960896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3928520679473877, "target_z": 0.32031246419191756, "ce": 5.982900142669678, "genuineness": 4.287685871124268, "identity": 0.2935887277126312, "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_TENS_reg_low/0812f93876d8c6844dc00a32e435f1e6266bac45b3a673cdc5b43e9b9cfd26e1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "08400852987e90b223bdeddad0e40e5880e87ba8ed2781ad77cc2e117a472888", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29c9330d99e2155a61dfd93f429065afee52f033ed8f1f32ffc6da51aeac1b2e", "generation_seconds_batch": 24.265701323049143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9504140615463257, "target_z": -0.3496093481439382, "ce": 6.1020402908325195, "genuineness": 4.400654315948486, "identity": 0.641402006149292, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/08400852987e90b223bdeddad0e40e5880e87ba8ed2781ad77cc2e117a472888.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "095d817df3b41a45765dda4a2cb24a9777d3346750a2f9b24928ee2131f2d860", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17d857d497b02d3ebcc784d879705b167b837627a6cf2b97cfef994a43ea732d", "generation_seconds_batch": 23.779829851002432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9617929458618164, "target_z": -0.36328130371212364, "ce": 6.268836498260498, "genuineness": 4.18099308013916, "identity": 0.7090294361114502, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/095d817df3b41a45765dda4a2cb24a9777d3346750a2f9b24928ee2131f2d860.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "09ac40a429f2848f4deaf1a6a06a1c150f3c45cd17d765c97e82cc9e302665ca", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7ea753024e56a0f132bedf0e7f25ec2663510979a156694f818e51fd227f73c", "generation_seconds_batch": 23.100803704932332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6360743045806885, "target_z": 0.028076114805868628, "ce": 6.38003396987915, "genuineness": 4.111956596374512, "identity": 0.5971614122390747, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/09ac40a429f2848f4deaf1a6a06a1c150f3c45cd17d765c97e82cc9e302665ca.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": "09f3f742a2fa37567623dbafff1e3bd147e62c1aa9d675faa5ff5ae9948a2482", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "8cbabfc7313626b9ce6560a0928db18d8246cd18fbfae9eed14a0989d7d61d57", "generation_seconds_batch": 25.011490056989715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436741828918457, "target_z": 0.26757804443181454, "ce": 6.141753673553467, "genuineness": 4.595211505889893, "identity": 0.7848949432373047, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/09f3f742a2fa37567623dbafff1e3bd147e62c1aa9d675faa5ff5ae9948a2482.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "0a06dd084c94267f0ee1efccd88505bb2d1f486500e8bd5ac6022090c9bacedd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71851b892641bafa78a6fc22043779e4a9772ac5c336f26a2fcd2ff0056ee781", "generation_seconds_batch": 25.603557397960685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.657815933227539, "target_z": 0.0019531160479793928, "ce": 6.197352409362793, "genuineness": 4.68307638168335, "identity": 0.679467499256134, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0a06dd084c94267f0ee1efccd88505bb2d1f486500e8bd5ac6022090c9bacedd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "0a2db864e11bf29423d671c8d6a445b456dd9cb012a8c4f5599d4891a89cf7b7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72109cf78aac2883d0d004dba0b621ee69e594703012736d6d2f64e0308b6bf0", "generation_seconds_batch": 24.08899778907653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5367127656936646, "target_z": 0.14746097778409273, "ce": 5.974957466125488, "genuineness": 4.300237655639648, "identity": 0.7123723030090332, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0a2db864e11bf29423d671c8d6a445b456dd9cb012a8c4f5599d4891a89cf7b7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "0a6cdf819cf4e5240b96250709bebca68e578f37a78a9df5ba0b7f8c9f7a9d27", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb3975a4fedc48883729f1ac5fd22a10d26b6d588f9ff321af84d5f1366e3bbd", "generation_seconds_batch": 24.16671350202523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8740134239196777, "target_z": -0.25781246419191756, "ce": 6.133810997009277, "genuineness": 4.406930446624756, "identity": 0.5634280443191528, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0a6cdf819cf4e5240b96250709bebca68e578f37a78a9df5ba0b7f8c9f7a9d27.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0b35fb93e43c1f119d3f0a10d794b275ef182eb32245ee36844819850737277d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "636c960854e433dfe469393d65d66c1a5c727d69eb4ec71b3fabf5c08df9eb61", "generation_seconds_batch": 17.07161507406272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6001091003417969, "target_z": 0.07128902221590727, "ce": 6.062326908111572, "genuineness": 4.645420074462891, "identity": 0.700982928276062, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0b35fb93e43c1f119d3f0a10d794b275ef182eb32245ee36844819850737277d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "0b41aa2cea04f53d359136ffa49a6abef35a4f354b16501a9eb244179a9add3e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ed931da214568f824402680f816dbb2f7a2e689b0d20627b496749e2b8b57541", "generation_seconds_batch": 23.486624783021398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5058274269104004, "target_z": 0.1845703259280309, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.8351042866706848, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/0b41aa2cea04f53d359136ffa49a6abef35a4f354b16501a9eb244179a9add3e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "0b691a4902509575aa15208e51c9a26fb43c908418e69993bde3e8a06940f734", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3068bc77ba00a710c7b37828accc4b40a593114e537e75222518876acdcfcab", "generation_seconds_batch": 23.81093512394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8398770093917847, "target_z": -0.2167968839520206, "ce": 6.086154937744141, "genuineness": 4.557555198669434, "identity": 0.8148159384727478, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in komen.", "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_TENS_reg_low/0b691a4902509575aa15208e51c9a26fb43c908418e69993bde3e8a06940f734.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0b6ff3bb73ab63e405abb891a7168179665c819aa04d86dac758a1ef650db0c5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbd2f8b21e77820930f3fe15f790cc27f0407c342a0f5c160da4e919e564e5a3", "generation_seconds_batch": 18.880745266913436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4107329845428467, "target_z": 0.29882818766414426, "ce": 5.982900142669678, "genuineness": 4.68307638168335, "identity": 0.5938059091567993, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0b6ff3bb73ab63e405abb891a7168179665c819aa04d86dac758a1ef650db0c5.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": "0ba210399d92774a520bd65894ce47aaaef907b4f3bfa6601e22cb786f19cfcd", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "be37a8fcebedf77164b2c0250c4594bcbd4631bc45f2f4928996074b9905fd92", "generation_seconds_batch": 21.8596139929723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4611248970031738, "target_z": 0.23828130371212364, "ce": 6.149696350097656, "genuineness": 4.796044826507568, "identity": 0.7577347755432129, "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_TENS_reg_low/0ba210399d92774a520bd65894ce47aaaef907b4f3bfa6601e22cb786f19cfcd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "0bb6320e2bcbc37eaf860c1a270cfe87347be7bf36dd37f2919133cb9daf52ee", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea7a40079d5d6a40f62902877249ef2739c2ae51e4915fce808ed8de2eb5190e", "generation_seconds_batch": 22.441705646924675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6917492151260376, "target_z": -0.038818396302085, "ce": 6.4515180587768555, "genuineness": 4.20609712600708, "identity": 0.3279643654823303, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0bb6320e2bcbc37eaf860c1a270cfe87347be7bf36dd37f2919133cb9daf52ee.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "0bdec03ef39d5fc355d14d9500ea9350f0d717ba8de4294a460b72fa43767ebe", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c437548aff536bdd91216e686401e1d9e5ff5ed2945f4227965ac8ea3b6f96e0", "generation_seconds_batch": 24.518208281020634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6647244691848755, "target_z": -0.006347627155933027, "ce": 6.181467056274414, "genuineness": 4.582659721374512, "identity": 0.5303078889846802, "hypothesis": "De deuren staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0bdec03ef39d5fc355d14d9500ea9350f0d717ba8de4294a460b72fa43767ebe.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0c25d3836c185f663648924d43225e6ba3a374319796c427f077e4f5db8a1bf0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7631c32979f0090f818c980268988ff0f3e26148e546ffc003a21e0b551b8f9", "generation_seconds_batch": 24.2920363729354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1165093183517456, "target_z": 0.6523437679040412, "ce": 6.149696350097656, "genuineness": 4.720732688903809, "identity": 0.7292853593826294, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0c25d3836c185f663648924d43225e6ba3a374319796c427f077e4f5db8a1bf0.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": "0c4067de7a72614fb0068f6649bd6f4cc551b6b835141a758fadbeb651a7ad37", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9801bc224117d3a4f535e5a3b4456957b603ec8e056a2593c086a29be8d8ad9", "generation_seconds_batch": 18.283076160936616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3749710321426392, "target_z": 0.3417968839520206, "ce": 6.006728172302246, "genuineness": 4.519899368286133, "identity": 0.647398829460144, "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_TENS_reg_low/0c4067de7a72614fb0068f6649bd6f4cc551b6b835141a758fadbeb651a7ad37.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": "0c4802052d0f4e3d86d725b49e59665c69a5ea701a1919af9c97991de999a5ab", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "9c30eaa4d9bb9aa68d8efe4714278372abb47e9f7beaf0915f9ad458bf9d3d6c", "generation_seconds_batch": 21.73340437200386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6423732042312622, "target_z": 0.02050786173611334, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.819391131401062, "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_TENS_reg_low/0c4802052d0f4e3d86d725b49e59665c69a5ea701a1919af9c97991de999a5ab.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very relaxed and unstrained. 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": "38d25a9f92417795f4b6ac3c00e54ec60749ae99639516b99f66b397d63c12c7", "generation_seconds_batch": 15.792407928034663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1731138229370117, "target_z": -0.6171875358080824, "ce": 6.125868320465088, "genuineness": 4.055472373962402, "identity": 0.6652992963790894, "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_TENS_reg_low/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "0cf84de38c5395c2271bf2289eea717519a7a0a6e89c90abe46ca33d8027d060", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "918c2611e0535d3235f566cd6899122ec105d3ff7c362b540d3621fbe378706b", "generation_seconds_batch": 18.090922074974515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3847243785858154, "target_z": 0.33007804443181454, "ce": 5.77639102935791, "genuineness": 4.532451152801514, "identity": 0.6279813051223755, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0cf84de38c5395c2271bf2289eea717519a7a0a6e89c90abe46ca33d8027d060.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "0d2e997d4220d585528c385d0fc00204076bd98f61aac447be29adafb979bbfb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "544b1f9123da59f1dd4f2d3fa478597e087b5ca89afc3bcc16770d3a79dbc021", "generation_seconds_batch": 18.967712357989512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5708492994308472, "target_z": 0.10644525431186605, "ce": 6.070269584655762, "genuineness": 4.256305694580078, "identity": 0.7780994176864624, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0d2e997d4220d585528c385d0fc00204076bd98f61aac447be29adafb979bbfb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "0d6409b5cd00ae82e5c3ac6d34084c5fcae7d4d7069dc0c4c2668659fead51c2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "bdbb64509951a933c176a99e43779ce43e7b04451a80fed4ded34b903a000205", "generation_seconds_batch": 18.178208232973702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0040571689605713, "target_z": -0.4140626074242473, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.6318303346633911, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0d6409b5cd00ae82e5c3ac6d34084c5fcae7d4d7069dc0c4c2668659fead51c2.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": "0d6643743e0a44cd5d38674837f865b6c4a79daf6d6db0a62d524f9e9ccb3fd5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb83427ad1f3bd327621619e400ec13e0310c62197e544400d89744fb0e73b69", "generation_seconds_batch": 20.661038902006112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5854791402816772, "target_z": 0.08886720988005152, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.6618481874465942, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0d6643743e0a44cd5d38674837f865b6c4a79daf6d6db0a62d524f9e9ccb3fd5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "0e0d1e5826e526d07692e3743f86c2f18b57893acbc3a7ebce59cb3a440cdb64", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29e34c13abbc1db0a6cf1c34ee2094e2b6019185c1758d262a3afe0ce971f286", "generation_seconds_batch": 17.484357842011377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3522133827209473, "target_z": 0.3691406518560618, "ce": 6.300607204437256, "genuineness": 4.375550270080566, "identity": 0.2900785207748413, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0e0d1e5826e526d07692e3743f86c2f18b57893acbc3a7ebce59cb3a440cdb64.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "0e3e56c231a7dd995710b03187980e55e18118609c6fcdf3febeedb98c0449c8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d2f1286a70fe6d850b6310445a88b311a466c054fd8cf889d3ac75ffdb2f857", "generation_seconds_batch": 24.518208281020634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127598524093628, "target_z": -0.5625, "ce": 6.038498878479004, "genuineness": 4.632867813110352, "identity": 0.5269689559936523, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0e3e56c231a7dd995710b03187980e55e18118609c6fcdf3febeedb98c0449c8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "0e40878f87b738383642667e121458a3a63afa7b87fc8af59608b148dfae990b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e715d5b117c3d61fecf8aaee7c6b2c81f600018040a35c54c1951fafb21edd21", "generation_seconds_batch": 17.950093136983924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.771604061126709, "target_z": -0.13476558023989696, "ce": 6.300607204437256, "genuineness": 4.068024158477783, "identity": 0.29055553674697876, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0e40878f87b738383642667e121458a3a63afa7b87fc8af59608b148dfae990b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "0ea47aa2de79bbe3745f9d88dcf9465a6e4ef98227185df663530ad414439e70", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "b293c6adf9813a01e15e78a5510c52ff855721ba733566c5c9766fce996cf8c9", "generation_seconds_batch": 19.49372464697808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0788321495056152, "target_z": -0.5039062320959588, "ce": 6.2767791748046875, "genuineness": 4.331617832183838, "identity": 0.5613403916358948, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0ea47aa2de79bbe3745f9d88dcf9465a6e4ef98227185df663530ad414439e70.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "0ee50c15a2480e8d2f5cfa0440a7180f8a93116308d03059a433259bde704e49", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "cd26abb2934c43c18346116624ad3e5a1e275cacdc5ddd650c8db73c1780cd29", "generation_seconds_batch": 22.493943006033078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2010376453399658, "target_z": 0.5507813037121236, "ce": 5.728734970092773, "genuineness": 4.413206577301025, "identity": 0.6009310483932495, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "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_TENS_reg_low/0ee50c15a2480e8d2f5cfa0440a7180f8a93116308d03059a433259bde704e49.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "0f751ebb0af81f845dffae5fa0ace9329953618a9fc940113bd9dbec7535f0c7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "123b8a786ca9cdad15f38860e180009f733b9af27918b9f0b9cf4167bdd61b3b", "generation_seconds_batch": 23.50184205302503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511980056762695, "target_z": -0.4707032592803091, "ce": 6.332377910614014, "genuineness": 4.243753433227539, "identity": 0.45967376232147217, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/0f751ebb0af81f845dffae5fa0ace9329953618a9fc940113bd9dbec7535f0c7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "0faaded1a129636a21ed5d073b520d1b939e561c0ad6552104cc8db5403e7832", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1375c72c5162fe31c4e2ac26bd9d90f86ea175fca6b2e9fdda680e2a4865640a", "generation_seconds_batch": 24.001254394999705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.673258662223816, "target_z": -0.016601629640154554, "ce": 5.990842819213867, "genuineness": 4.124508857727051, "identity": 0.6249635219573975, "hypothesis": "Ik 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_TENS_reg_low/0faaded1a129636a21ed5d073b520d1b939e561c0ad6552104cc8db5403e7832.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "1043d3c70fbe5dd4d2a652dd74af8c8b08a8f5fc9de670ec934c015afda961f8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "14a298dc8ec3cf4cad4a91dae6b5fe7f87ccc8a3c886ffbd8222e5563525e347", "generation_seconds_batch": 17.819917223067023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.064202308654785, "target_z": -0.48632818766414426, "ce": 6.133810997009277, "genuineness": 4.105680465698242, "identity": 0.6108158826828003, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1043d3c70fbe5dd4d2a652dd74af8c8b08a8f5fc9de670ec934c015afda961f8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "107bb4216a339c7561260797f876624cd8494d3964712276031025b53443275b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e888627ff91cd90949a66dea078b98027e917cf90ba5df1014ebffd9b4d2d336", "generation_seconds_batch": 17.63042104500346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746814727783203, "target_z": -0.8593750716161649, "ce": 6.141753673553467, "genuineness": 4.193545341491699, "identity": 0.6634311676025391, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/107bb4216a339c7561260797f876624cd8494d3964712276031025b53443275b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "10d84623c38febf8c1c27e6d0ade0b286f2e14dbc181aa2f581734fefac14b5f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f94136d0cbd69d3dccf62d400c55b067bb4cb9e525d70b6a7e3868d8577345e", "generation_seconds_batch": 16.57739093794953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.714709997177124, "target_z": -0.06640623209595879, "ce": 5.847874641418457, "genuineness": 4.093128204345703, "identity": 0.4699556827545166, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/10d84623c38febf8c1c27e6d0ade0b286f2e14dbc181aa2f581734fefac14b5f.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": "10f0a4e499a10f443b36f2578886bacf7ce0fd2d402171e880ae3765f5029771", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "1817f855b9a5497974531b9b0db3ef002243875166bdf2d0cbbb834bcc6985fa", "generation_seconds_batch": 20.19667174795177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.091836452484131, "target_z": -0.519531160479794, "ce": 6.3641486167907715, "genuineness": 4.519899368286133, "identity": 0.567955493927002, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/10f0a4e499a10f443b36f2578886bacf7ce0fd2d402171e880ae3765f5029771.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "117a15bea9f05201033853d56c8599533c30745ee904bd02d29041fe38e3288f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5d9e9fd66871bf034ae7c35c90b1a66c895b026405f905b51287672e64f9792", "generation_seconds_batch": 26.947682947968133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5245212316513062, "target_z": 0.16210934814393818, "ce": 5.80021858215332, "genuineness": 4.507347106933594, "identity": 0.4536769986152649, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/117a15bea9f05201033853d56c8599533c30745ee904bd02d29041fe38e3288f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "121bb6e742728f11e27a80b2a0b6aa2a338162b5fcdb08f6d80e57d4f2160d60", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e15e1e3148bd646beaf4e6c120cd891ee4c3de6fe4d61ab78ea2734002dca822", "generation_seconds_batch": 16.892138245049864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.746814489364624, "target_z": -0.1049804172758815, "ce": 6.3641486167907715, "genuineness": 4.545003414154053, "identity": 0.38448914885520935, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/121bb6e742728f11e27a80b2a0b6aa2a338162b5fcdb08f6d80e57d4f2160d60.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "1222300709f082c5237d7a2f60c375ec9b049fa257e73d259d18d2b0e6723963", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "135e232dcf6579a9b2509976121fbaa57b97de1f92c03aac7c05679fcef71309", "generation_seconds_batch": 23.71445237705484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3262046575546265, "target_z": 0.4003906518560618, "ce": 6.292664527893066, "genuineness": 4.494794845581055, "identity": 0.7057569622993469, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1222300709f082c5237d7a2f60c375ec9b049fa257e73d259d18d2b0e6723963.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "122e3dedfadcdbfd822d0aa7036f93cfda3616b096ac7d3cb19ba0966a42ad7d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "8194a0a591034bf6f52d75cee20846cc2c0f5281f8705e1f25624fe99b11d82b", "generation_seconds_batch": 27.730835248017684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1750290393829346, "target_z": 0.582031160479794, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.7859224081039429, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/122e3dedfadcdbfd822d0aa7036f93cfda3616b096ac7d3cb19ba0966a42ad7d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "125ec06a5f01a9fb91fbedf6e6aab5708a455da0939def1f04b4df5b64fbcd38", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e91d91ec6874eacd579d0847892d7f9c796a1869f7a11acc9fa44e333c3bdb6c", "generation_seconds_batch": 26.09549324598629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1100071668624878, "target_z": 0.6601562320959588, "ce": 5.784333229064941, "genuineness": 4.532451152801514, "identity": 0.6701458096504211, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/125ec06a5f01a9fb91fbedf6e6aab5708a455da0939def1f04b4df5b64fbcd38.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "1282c3afbb329403f2221786084facf73c227dab1f9f6354dadade6e537509d0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4bec01dd7f440a6390eac1b0f14cdbfbbf544a6593c2eec17066e36635e10d1", "generation_seconds_batch": 26.839735769899562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2530550956726074, "target_z": 0.48828130371212364, "ce": 5.959072113037109, "genuineness": 4.400654315948486, "identity": 0.5199947357177734, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1282c3afbb329403f2221786084facf73c227dab1f9f6354dadade6e537509d0.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very relaxed and unstrained. 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": "9882205c79ed3cf0ebcfaa194be3604052bec5c9042c9f3ef79e2ce2465ca9c6", "generation_seconds_batch": 17.145708944066428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.034942388534546, "target_z": -0.45117181233585574, "ce": 6.022613525390625, "genuineness": 4.444586753845215, "identity": 0.6873957514762878, "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_TENS_reg_low/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "12f83a176088a3b459763337a04e8f2086b8e529d1595192117f8a8407dac72b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "2300be3157fcacff357b1f334bcf2d66391781757b9fe61a89067f0e1df6a6a6", "generation_seconds_batch": 16.945996257010847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6131134033203125, "target_z": 0.055664093832072124, "ce": 6.23706579208374, "genuineness": 4.582659721374512, "identity": 0.7236504554748535, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/12f83a176088a3b459763337a04e8f2086b8e529d1595192117f8a8407dac72b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "13186ea61ccfceb1abf6600fa6bcfad9a6645c2ecf7aba25786c33eb483894f4", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9d1d9e7ed629182967416b4217135e09aab5fe259ae3473f6f553cf1acfda7ce", "generation_seconds_batch": 20.43949282192625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5163934230804443, "target_z": 0.17187507161616486, "ce": 6.443574905395508, "genuineness": 4.413206577301025, "identity": 0.434212863445282, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "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_TENS_reg_low/13186ea61ccfceb1abf6600fa6bcfad9a6645c2ecf7aba25786c33eb483894f4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "138c7f23ba91ee13216d65215807b52eab320b6ead353c9bfb99193e45ad9fcd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a75f265df550493b3afb39d64ab4255f4f37f6fa1447af00125a173651f3c45", "generation_seconds_batch": 17.698533938033506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.12109637260437, "target_z": -0.5546875358080824, "ce": 6.221180438995361, "genuineness": 4.607763767242432, "identity": 0.6927107572555542, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/138c7f23ba91ee13216d65215807b52eab320b6ead353c9bfb99193e45ad9fcd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "13b2c639991490629f18ff3baac23ef2f3eb9666b91322a4b1086c3b783d3586", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d7129e871a5e512575881e09d39d53af0f01ffc2b1476d742209792c1597fb6e", "generation_seconds_batch": 17.27087054902222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5098912715911865, "target_z": 0.17968753580808242, "ce": 6.340320587158203, "genuineness": 4.507347106933594, "identity": 0.41002053022384644, "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_TENS_reg_low/13b2c639991490629f18ff3baac23ef2f3eb9666b91322a4b1086c3b783d3586.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "13b80968038fd805533d751ec6d5846d201beed4de8f3cf0d0331c3194ab1928", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4ddf934763cff99085505bbbbbfe474047bdaf273e06e4a7178ed4833e9f15e", "generation_seconds_batch": 25.345918715931475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595572471618652, "target_z": 0.4804688395202061, "ce": 5.780362129211426, "genuineness": 4.356721878051758, "identity": 0.7483937740325928, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/13b80968038fd805533d751ec6d5846d201beed4de8f3cf0d0331c3194ab1928.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "1441e1cdb4501190a7eca1a46a8f9425422846116e3b2aff033a29e04ce6106b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11b6255e9febf27040b4a636caef60633cbbeeb10431c751763e85ee4271d063", "generation_seconds_batch": 23.509477789979428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5424022674560547, "target_z": 0.14062492838383514, "ce": 6.356205940246582, "genuineness": 4.532451152801514, "identity": 0.35250598192214966, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1441e1cdb4501190a7eca1a46a8f9425422846116e3b2aff033a29e04ce6106b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "14a66a3f96375c5125ac1477154b4900c7abfb2f20e2ad80b166b6f21e77bd7c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cae2834f4cb7c900eac7825882b9b48c432f5e52c41d12e164ad19e4b89316b0", "generation_seconds_batch": 25.155860129976645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6954066753387451, "target_z": -0.04321290741003864, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.7276554703712463, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/14a66a3f96375c5125ac1477154b4900c7abfb2f20e2ad80b166b6f21e77bd7c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "15294ce8cdfa65039ee678c98468c19c04c8b0bda5ad93f157efe4f6136e4ea9", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "323f023353f69aa35f1c3f3f000110675f8ba4f56de917f96f6d3b9320d9c966", "generation_seconds_batch": 17.912434212048538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.020312547683716, "target_z": -0.4335937679040412, "ce": 6.229123115539551, "genuineness": 4.3881025314331055, "identity": 0.48303085565567017, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/15294ce8cdfa65039ee678c98468c19c04c8b0bda5ad93f157efe4f6136e4ea9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "154a553cb15ac73e651ba118d039851d3980f782debe173d489a0e0fe7b8a90c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "51562f7274d713439408c5b9604f55df9a7dd7ea2e722adbb7202d53122770be", "generation_seconds_batch": 26.658779768040404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.230297565460205, "target_z": 0.5156249283838351, "ce": 5.998785495758057, "genuineness": 4.7332844734191895, "identity": 0.733709454536438, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. We've regarded ensemble.", "wer": 0.3157894736842105, "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_TENS_reg_low/154a553cb15ac73e651ba118d039851d3980f782debe173d489a0e0fe7b8a90c.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very relaxed and unstrained. 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": "f178b51cabb3d453104d8aee20ac3f30d866c020590a2f57c80a44b646b7ec8d", "generation_seconds_batch": 18.05504199594725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4952614307403564, "target_z": 0.19726558023989696, "ce": 6.395919322967529, "genuineness": 4.394378662109375, "identity": 0.5181436538696289, "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_TENS_reg_low/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.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": "15abd01856462d4a6861abe31c3a8910dcbbbd2e24dda0add8f38d8d139dd5a0", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "53b1a324400981aee8cccd86beefaf34f951755793b38282deabac95b1ee0a3d", "generation_seconds_batch": 20.717880478012376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7764806747436523, "target_z": -0.14062492838383514, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.6238820552825928, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en leg mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/15abd01856462d4a6861abe31c3a8910dcbbbd2e24dda0add8f38d8d139dd5a0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "15ca4e6c646359a83da82c730365b91df26060d1378b9ef480623055da63e53f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9178d020b2e15a219eafffc45f3e7a38bfe6f03d643a8075970eec9ffc104dba", "generation_seconds_batch": 22.053278245963156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5367127656936646, "target_z": 0.14746097778409273, "ce": 5.871703147888184, "genuineness": 4.808597087860107, "identity": 0.8131580352783203, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/15ca4e6c646359a83da82c730365b91df26060d1378b9ef480623055da63e53f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "15dc6e1e6e61c3603ea680197b7bbcbcbaf46bc19dd570999198307ab36fe5fc", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a99dbe2a95cd103931c2ad6239bda4a16bbf9e8a02e512f93b1c943fb28732ae", "generation_seconds_batch": 16.945996257010847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4562482833862305, "target_z": 0.24414065185606182, "ce": 6.086154937744141, "genuineness": 4.419482707977295, "identity": 0.748013436794281, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/15dc6e1e6e61c3603ea680197b7bbcbcbaf46bc19dd570999198307ab36fe5fc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "15df7aaf1e0df9d649979fc0c3bf941416bfb4471f71e1937e00b5cb35ff9c15", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "24b6c79e0c46e6bb7825c2e709671119678c92d158b3209872603421a5c28640", "generation_seconds_batch": 18.06196842691861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8301236629486084, "target_z": -0.20507804443181454, "ce": 6.38003396987915, "genuineness": 4.595211505889893, "identity": 0.18658465147018433, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/15df7aaf1e0df9d649979fc0c3bf941416bfb4471f71e1937e00b5cb35ff9c15.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "161d39b1f837b388b5657481fd1b44ad6c8ce45c937cd33276caaf4f9ab1a50e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e8dfbcdf1016d68797831c8b2e7922bca75b927a3c417fc79f6208c927732ea", "generation_seconds_batch": 18.744288856978528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5903557538986206, "target_z": 0.08300786173611334, "ce": 5.855817794799805, "genuineness": 4.174716949462891, "identity": 0.27970728278160095, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/161d39b1f837b388b5657481fd1b44ad6c8ce45c937cd33276caaf4f9ab1a50e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "1630a3a717ce37108b9ba5e97a669d9a16f8ed9269e41ca2485df5abcf401323", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7d337e8080e995e190a2fcc14f16deacf934c92f6fe94cd8f97264c6a112887", "generation_seconds_batch": 24.85925367800519, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8382514715194702, "target_z": -0.21484376790404122, "ce": 6.133810997009277, "genuineness": 4.406930446624756, "identity": 0.42995721101760864, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1630a3a717ce37108b9ba5e97a669d9a16f8ed9269e41ca2485df5abcf401323.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "16613109c197fbe774013282c8ad98ba330f065e536af1bbbebc13d3246a90ef", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6702f603b4861cf1457e2d048d6991127d0c40836cb54f4d468a0ca200cd168", "generation_seconds_batch": 23.11839755496476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6781352758407593, "target_z": -0.022460977784092732, "ce": 6.356205940246582, "genuineness": 4.770940780639648, "identity": 0.6933654546737671, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/16613109c197fbe774013282c8ad98ba330f065e536af1bbbebc13d3246a90ef.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "1709ab5b08980eb7e8743f7208eb53a84ad06ab5bc988f9b34aba75a6ae4619d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "49b17ccdca0cec42f180760bee5311b859b088e509a996232aa2d42916882b1c", "generation_seconds_batch": 16.350950531079434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7476272583007812, "target_z": -0.1059569752998712, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.8061281442642212, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1709ab5b08980eb7e8743f7208eb53a84ad06ab5bc988f9b34aba75a6ae4619d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "172b44519c02160e0f8626c524710c193c85f8c51eff829ad2b74f994d5bcc70", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebe3680f623972c89a4d5b89bf69d953170b6f5b4585aa1a0dd9c0a44192000f", "generation_seconds_batch": 27.552371879923157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3148258924484253, "target_z": 0.41406246419191756, "ce": 5.919358730316162, "genuineness": 4.137060642242432, "identity": 0.3437000513076782, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/172b44519c02160e0f8626c524710c193c85f8c51eff829ad2b74f994d5bcc70.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "175c8215370a8d9d550c4aec986c3d327345014425a2ca5d73d924e7769a5866", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c220b55a673b721e202c302d96b82be1bec9f034844f8def14a6ce1f65d8bf0", "generation_seconds_batch": 23.769813135964796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6476563215255737, "target_z": 0.014160091347850599, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.5828162431716919, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/175c8215370a8d9d550c4aec986c3d327345014425a2ca5d73d924e7769a5866.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "176f6d8751297c3ce9ff98a6ff2aeffacbcec8405433ded9d4c19f6732277259", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "522218175bbcd50c8037f0cdbc44316a32fe7031414bb25978555ba487743f3b", "generation_seconds_batch": 24.260291144018993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.607424020767212, "target_z": 0.0625, "ce": 5.752562999725342, "genuineness": 4.381826400756836, "identity": 0.3362250030040741, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/176f6d8751297c3ce9ff98a6ff2aeffacbcec8405433ded9d4c19f6732277259.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "17c8d4be04b225462f47ebf58b4537e764164f980cf9e64f7e24b6674437bd02", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "3efccba4b6383219049df4c974a588689121e54643daa12ca60511e4fe3c5f38", "generation_seconds_batch": 18.982315422967076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.61819326877594, "target_z": 0.04956053456597167, "ce": 6.443574905395508, "genuineness": 4.231201171875, "identity": 0.5987372398376465, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/17c8d4be04b225462f47ebf58b4537e764164f980cf9e64f7e24b6674437bd02.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "17eec6f72bf94f3a15a7750c954d0a2e2ab796046168c6e9e959b6560f7507bf", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a71858bc3df8f70f809249bdde72d172e948079f03dadd51e033578d6032193e", "generation_seconds_batch": 18.693622352089733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1015896797180176, "target_z": -0.5312498567676703, "ce": 6.252951145172119, "genuineness": 4.281409740447998, "identity": 0.707119345664978, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/17eec6f72bf94f3a15a7750c954d0a2e2ab796046168c6e9e959b6560f7507bf.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": "17fa28f540f243d34f059d2084a09eeb08e3f5a890d41e67c58318839c4ff269", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f76227e0d49b6b059a44647abdb294dca494ee5a95da20d78d28202413d7b028", "generation_seconds_batch": 23.833263900014572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0235636234283447, "target_z": -0.4375, "ce": 6.260893821716309, "genuineness": 4.545003414154053, "identity": 0.7969968318939209, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/17fa28f540f243d34f059d2084a09eeb08e3f5a890d41e67c58318839c4ff269.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "17fffba827496ee78dd4f59c3ba4c45a66be1465ccf6b18bd56ebe2384bc73ad", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d062afb20ca1a991f78d297e5fc8e724b446b59d63329982e356a23f1de7f730", "generation_seconds_batch": 24.120508004911244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.383098840713501, "target_z": 0.3320311604797939, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.785378098487854, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/17fffba827496ee78dd4f59c3ba4c45a66be1465ccf6b18bd56ebe2384bc73ad.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "183034ae3c201daa81f93c97ea98edace6242afbb11d03fc66bab1a74ede842c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8dea11081c326f0d9f39faf302532459c124e94571e53085e6e7e6be3f38979", "generation_seconds_batch": 19.296586651937105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.043070077896118, "target_z": -0.4609373925757527, "ce": 6.157639026641846, "genuineness": 4.149612903594971, "identity": 0.271165668964386, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/183034ae3c201daa81f93c97ea98edace6242afbb11d03fc66bab1a74ede842c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "18dbcfb00c503749ba82e0c4f2a7388a5b6e5c1602cf34d4759aacc93e796797", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fbeb7ea314f4ddec5b988a70c3e7be945c6cbc27fa1ffeaf8ef8d41e1fdc110", "generation_seconds_batch": 25.2665543720359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6082367897033691, "target_z": 0.061523441976010305, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.7109977006912231, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/18dbcfb00c503749ba82e0c4f2a7388a5b6e5c1602cf34d4759aacc93e796797.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "19567f6ba4ee03be804677b7e1258b3702376fba93a877a503ef8325999ae2b1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "770cd6816e0b05d1e2fdbe4f1fb985659af2b198719d9c3ddfb9f1a04479b1b8", "generation_seconds_batch": 18.29483273893129, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3099491596221924, "target_z": 0.41992195556818546, "ce": 5.9431867599487305, "genuineness": 4.65797233581543, "identity": 0.6843794584274292, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/19567f6ba4ee03be804677b7e1258b3702376fba93a877a503ef8325999ae2b1.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": "19fa972db7f8982b83961dab69a4a326b1d60059e268d138876ad724682330d2", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5695f158871a35c9f2416124ccd3edfd62c11dcc120f131e2c8130061492a648", "generation_seconds_batch": 15.864832510938868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6897172927856445, "target_z": -0.03637700124211076, "ce": 6.38797664642334, "genuineness": 4.394378662109375, "identity": 0.2005978524684906, "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_TENS_reg_low/19fa972db7f8982b83961dab69a4a326b1d60059e268d138876ad724682330d2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "19fcb8d5ab095a590f3ed63c5fee04b399a4a985913688a93de6e6f77f38d83d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ff2fd36970185b2219facef63700f3118daaca3516fc8fde65c66ffee7aba9fc", "generation_seconds_batch": 18.723675807006657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.082083225250244, "target_z": -0.5078124641919176, "ce": 6.316492557525635, "genuineness": 4.18726921081543, "identity": 0.5028839707374573, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/19fcb8d5ab095a590f3ed63c5fee04b399a4a985913688a93de6e6f77f38d83d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "19ff67a499f646250cec6e9b3116cfdc5ee4f7818113faa5b8006b51898a98bb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2be1cc4bda770da31d806c7ef8186a33fbb628b1aa8a6972a8985fbb0d3d30c", "generation_seconds_batch": 17.07161507406272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4700654745101929, "target_z": 0.22753902221590727, "ce": 6.181467056274414, "genuineness": 4.519899368286133, "identity": 0.6973965764045715, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/19ff67a499f646250cec6e9b3116cfdc5ee4f7818113faa5b8006b51898a98bb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "1ac25643f7946bb653a636484723008e23f45fc8f7d8057823ab9fb06559b214", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c9c40fa2858984f20613f40e1ccdd8deed6a81419064ce36343917dd7aea14d", "generation_seconds_batch": 24.343175414949656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3733454942703247, "target_z": 0.34375, "ce": 6.070269584655762, "genuineness": 4.413206577301025, "identity": 0.7362771034240723, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1ac25643f7946bb653a636484723008e23f45fc8f7d8057823ab9fb06559b214.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "1b9842f6cd9b071e72246e9d4a8e5039433ad83025919adddbce101612e83158", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "348b1882f7f837315dfd180b0f902843d4f9e45899f2a0667f13d61b31e4f8eb", "generation_seconds_batch": 18.723675807006657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.002431631088257, "target_z": -0.4121094913762679, "ce": 6.221180438995361, "genuineness": 4.413206577301025, "identity": 0.6372218728065491, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1b9842f6cd9b071e72246e9d4a8e5039433ad83025919adddbce101612e83158.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "1baa861940d0bb801dad487b75d3ffe825639437ccbd89ab95785fafb0a8678a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8d581f7cfae030f836bbb6bd79131562837f97c8443eb132f6466f758a702de", "generation_seconds_batch": 23.43494895193726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.504201889038086, "target_z": 0.1865234419760103, "ce": 6.213237762451172, "genuineness": 4.300237655639648, "identity": 0.6249725818634033, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1baa861940d0bb801dad487b75d3ffe825639437ccbd89ab95785fafb0a8678a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "1c23129a61f1ce268e0f7d5ddf605b0fa24b9e47692c726a60f96fb89224ab72", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32e2b88fdf0c0d056364acd5e6a7dd16031374e943f94bdf1e4c66b4825f8355", "generation_seconds_batch": 25.66421750409063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7545359134674072, "target_z": -0.11425786173611334, "ce": 5.561938762664795, "genuineness": 4.337893962860107, "identity": 0.48150962591171265, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1c23129a61f1ce268e0f7d5ddf605b0fa24b9e47692c726a60f96fb89224ab72.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "1c49a5b9917b315b521f73f41c8f2a8f94bd91e5bd664d47791a977254a5fd91", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2309d71d185d2070a47f05ec611812562940d37d6e2c62f62a4caaacb9b951d", "generation_seconds_batch": 24.43980036000721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4529972076416016, "target_z": 0.2480468839520206, "ce": 6.348263263702393, "genuineness": 4.570107460021973, "identity": 0.42379650473594666, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1c49a5b9917b315b521f73f41c8f2a8f94bd91e5bd664d47791a977254a5fd91.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "1ca5d4bdce17d06a4819cf1cc0ad7a56cfe908e091a8ca0fa985120902e15bb1", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2ec2e8372c69a8f9de6c3fcac7e622f3578360dc365c2aaf32972ad655c5981d", "generation_seconds_batch": 18.33999709400814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8707623481750488, "target_z": -0.2539062320959588, "ce": 5.760505676269531, "genuineness": 4.036643981933594, "identity": 0.6557728052139282, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1ca5d4bdce17d06a4819cf1cc0ad7a56cfe908e091a8ca0fa985120902e15bb1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "1cca22e79fd14d73a39b2101b2e84e94e5a1090327cea61c54dce6040ab92c11", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "deff54ee1022d6f3dd075eecabbede683356c7fe0d51b5ad4701cbe1b2127cc7", "generation_seconds_batch": 19.180992110050283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4700654745101929, "target_z": 0.22753902221590727, "ce": 6.149696350097656, "genuineness": 4.21237325668335, "identity": 0.7820943593978882, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1cca22e79fd14d73a39b2101b2e84e94e5a1090327cea61c54dce6040ab92c11.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "1d390dd8b88053ea5667d2fd714f24454aebe29aae26113cc248b1d5234db436", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b4cb77f5ea2876eb1c213dcae0707dc86de7e6929aa19db347d8d315ee4c5d4e", "generation_seconds_batch": 24.334042980102822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4286140203475952, "target_z": 0.2773437679040412, "ce": 5.661221981048584, "genuineness": 4.256305694580078, "identity": 0.468650758266449, "hypothesis": "Die Tür 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_TENS_reg_low/1d390dd8b88053ea5667d2fd714f24454aebe29aae26113cc248b1d5234db436.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very relaxed and unstrained. 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": "68bd8a27495f4bc5834b54f108f170b9967dfb7b7079591dbbd149c880e3ddb7", "generation_seconds_batch": 15.347440009005368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.825247049331665, "target_z": -0.19921869628787636, "ce": 6.308549880981445, "genuineness": 4.419482707977295, "identity": 0.5716466903686523, "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_TENS_reg_low/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "1d5f4fc7761396d5a52f15ea34886f664b322b9dbf61561838b8bdabcfff30a9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8684a0fa39404c7d2b0ba584a5bc3d8142d5c08603d361b907ff6af83f2f1e5b", "generation_seconds_batch": 23.31476210197434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0398190021514893, "target_z": -0.4570311604797939, "ce": 6.1020402908325195, "genuineness": 4.65797233581543, "identity": 0.6780683398246765, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1d5f4fc7761396d5a52f15ea34886f664b322b9dbf61561838b8bdabcfff30a9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "1d6023bf0c748d64452b422e736d2c9f5cb41ac07fa27c3354ab850f6c2c00cc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "039cddbbcbd8fddc7b85ed0a48c0c99a9d39d57fb094cf308f45e4673ae3a9f9", "generation_seconds_batch": 17.924870752030984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1620246171951294, "target_z": 0.5976562320959588, "ce": 5.824047088623047, "genuineness": 4.256305694580078, "identity": 0.8239821195602417, "hypothesis": "La porte encore ouverte, un j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1d6023bf0c748d64452b422e736d2c9f5cb41ac07fa27c3354ab850f6c2c00cc.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very relaxed and unstrained. 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": "6ed9a98cce5297bd54d583abf4fd7c37a7553bf04d07358a86e2d3452dafe690", "generation_seconds_batch": 18.4237529640086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.697235345840454, "target_z": -0.045410091347850595, "ce": 6.24500846862793, "genuineness": 4.381826400756836, "identity": 0.6586456894874573, "hypothesis": "Die Schlüsse liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "1dab0ad2cead269365e0dcaa5f87e28542fc2c6dcf07fc60b579fe92478a4241", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe3e84327deecacc764179aa10cec6c65145a6c12ef5e32696c13ce7b4bf9dd0", "generation_seconds_batch": 17.996784696006216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3148258924484253, "target_z": 0.41406246419191756, "ce": 6.181467056274414, "genuineness": 4.507347106933594, "identity": 0.7332977652549744, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1dab0ad2cead269365e0dcaa5f87e28542fc2c6dcf07fc60b579fe92478a4241.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "1e17c299bc4d2670a8556a408f6022d63fe4e638630f7f60407bb9df4dcc715c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab45c7132f49094a0062663afa21daceac7b18741865cfe9d1bbf3f29b15175c", "generation_seconds_batch": 17.83852531807497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3570899963378906, "target_z": 0.36328130371212364, "ce": 5.919358730316162, "genuineness": 4.532451152801514, "identity": 0.6817946434020996, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "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_TENS_reg_low/1e17c299bc4d2670a8556a408f6022d63fe4e638630f7f60407bb9df4dcc715c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "1e605b7ce2adeb47e971784bbe3890eb252613f752e191568861bcfc8948135f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a33b23a59df72ff44dc649831abb0dbfb86c54e0488f85be820d10c189e1bc5", "generation_seconds_batch": 26.575422744033858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1035048961639404, "target_z": 0.667968839520206, "ce": 6.022613525390625, "genuineness": 4.243753433227539, "identity": 0.6898810863494873, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1e605b7ce2adeb47e971784bbe3890eb252613f752e191568861bcfc8948135f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "1ea86b3c43b959ca28e7b775ec5f8d658bc009acca2bc1b24a582904291024e1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15f02a6552ad1d6c21e02ad16001eaae80eb06244703cca28d6d0a50e4462457", "generation_seconds_batch": 18.182384665007703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4749420881271362, "target_z": 0.2216796740719691, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.7663394808769226, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1ea86b3c43b959ca28e7b775ec5f8d658bc009acca2bc1b24a582904291024e1.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very relaxed and unstrained. 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": "136d35e024652d393b89561dcbd2edcc6c9be34599f950df754d9b40841c0749", "generation_seconds_batch": 15.347440009005368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7382804155349731, "target_z": -0.0947265580239897, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.682593584060669, "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_TENS_reg_low/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.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": "1ee6c7489a6d947a7323c7cf35f5b720ab3ab4b562efcc8a717d5ef84995cdc3", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4afe12656ec7b7c990ebeb75004cd3cc403147e2191baadbe7aec19544f4625", "generation_seconds_batch": 23.833263900014572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.844753623008728, "target_z": -0.22265623209595878, "ce": 6.348263263702393, "genuineness": 4.645420074462891, "identity": 0.636157751083374, "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_TENS_reg_low/1ee6c7489a6d947a7323c7cf35f5b720ab3ab4b562efcc8a717d5ef84995cdc3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "1f3e5a521eb4f96d07e7dd7b922f12e6ecc8bc71c80f730522361ab8f2f7419e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec37ad7271a862cd29380bc28d840bf75c824da9ee40c81b1cd07ad765909624", "generation_seconds_batch": 25.35504110401962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.75372314453125, "target_z": -0.11328130371212364, "ce": 5.95112943649292, "genuineness": 4.118232727050781, "identity": 0.427866667509079, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1f3e5a521eb4f96d07e7dd7b922f12e6ecc8bc71c80f730522361ab8f2f7419e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "1f4cfcd8657637a8604424a65fd29d93d3bbe11041e78c35f3b4f996d8897f94", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "392b85ced13a7b3d7e7c913013870c567e94c1ca19d66b7c0e634b277a867759", "generation_seconds_batch": 22.336135156918317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4692527055740356, "target_z": 0.22851558023989696, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.6468662023544312, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1f4cfcd8657637a8604424a65fd29d93d3bbe11041e78c35f3b4f996d8897f94.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "1f620e2353bbdf116e3fd5ab4ffc20ee55ed7fdf1b68c312e8bb8b071d5c3033", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2776939efb33fb8304c839c35b1ade72561fa4129bcd6580a55d27f1b683dfe9", "generation_seconds_batch": 22.649227416026406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5940132141113281, "target_z": 0.0786133506281597, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.39085111021995544, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1f620e2353bbdf116e3fd5ab4ffc20ee55ed7fdf1b68c312e8bb8b071d5c3033.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "1f7e56e3ba2a15377c7df9f7dadd825f2bf5a8db117ca9dc4175622c06aa1eeb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37a797fe8cbce5d9837f7a77bd4786d3f73e10dcfc4432a188d2e661517401ce", "generation_seconds_batch": 26.697074832045473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6789480447769165, "target_z": -0.02343753580808243, "ce": 5.982900142669678, "genuineness": 4.595211505889893, "identity": 0.6725932359695435, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1f7e56e3ba2a15377c7df9f7dadd825f2bf5a8db117ca9dc4175622c06aa1eeb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "1fec176b5eadf4d770fa4179064e097c3631770a1a8dea421260d4cf1ba20b2b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6f6c9d873d5e27b745640727849b89490d6380c2b4e30f66b09a903e6a07362", "generation_seconds_batch": 23.389638917986304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1847822666168213, "target_z": 0.5703124641919176, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.765895426273346, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1fec176b5eadf4d770fa4179064e097c3631770a1a8dea421260d4cf1ba20b2b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "1fed183af496bbeaa685791acba5c61f170be8bd271be74934392c5cc5e1b956", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1361af4c30d3489da9139ee8170adecf439ceef3bb73be10b7c2767f04c8632d", "generation_seconds_batch": 23.15190843003802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4156097173690796, "target_z": 0.29296869628787636, "ce": 6.221180438995361, "genuineness": 4.293961524963379, "identity": 0.5991078019142151, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/1fed183af496bbeaa685791acba5c61f170be8bd271be74934392c5cc5e1b956.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "2040697c76e4461a9ab46f9b0a6ad1e71a872d4f4a583ab7f65c0975ed32c164", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe9140a5566f8ae5d09acd13661c1191c4741762b74e5037aaa9e30422596f8a", "generation_seconds_batch": 27.53181885892991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9260308742523193, "target_z": -0.32031246419191756, "ce": 6.324435234069824, "genuineness": 4.532451152801514, "identity": 0.35296282172203064, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2040697c76e4461a9ab46f9b0a6ad1e71a872d4f4a583ab7f65c0975ed32c164.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "20483af342ba1138e94dbdb9807a6be21278b7e03e8e78c30fe0484cd9fbf3fd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb426431921c69c96c8f0a9f7fcad0072e68004d9e9c69450326d56aea4ea124", "generation_seconds_batch": 26.66754546505399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9081499576568604, "target_z": -0.29882818766414426, "ce": 5.855817794799805, "genuineness": 4.319066047668457, "identity": 0.7307411432266235, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/20483af342ba1138e94dbdb9807a6be21278b7e03e8e78c30fe0484cd9fbf3fd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "208a7b885e60ee4cea4fb12d2b411068c8efcb51a06b053dee9456e93f8de06a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0fa6d5b9d2fb49b4e13a9258e511845b510360e05ae5f3e0ec5a099bf2fdc50", "generation_seconds_batch": 23.70311780902557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4107329845428467, "target_z": 0.29882818766414426, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.27179500460624695, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/208a7b885e60ee4cea4fb12d2b411068c8efcb51a06b053dee9456e93f8de06a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "20abda1104f14371f10f6cb3bca33a5f4d9656034125ac293391dd166f746fdb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "528cef680d8c56e0b6ea649c5fa3b611c8a8b77ee0220aa9d07938476680aca7", "generation_seconds_batch": 20.43949282192625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6159580945968628, "target_z": 0.05224614074810819, "ce": 6.3641486167907715, "genuineness": 4.168440818786621, "identity": 0.5409277677536011, "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_TENS_reg_low/20abda1104f14371f10f6cb3bca33a5f4d9656034125ac293391dd166f746fdb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "20bea37359277c9408b8d4e6fd89f298bcb49a6326462ca74d226da01a596cbb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd98ab635e004d2b868cfc37315c44d45db3aeeb4a79af8d835a2cf1dae236ca", "generation_seconds_batch": 23.064952155924402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7992383241653442, "target_z": -0.16796869628787636, "ce": 6.252951145172119, "genuineness": 4.432034492492676, "identity": 0.5584211945533752, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/20bea37359277c9408b8d4e6fd89f298bcb49a6326462ca74d226da01a596cbb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "218c40312ac1d2f607a057970ab23053e208833aaecb1e608c4f9fa1add775ad", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0319ae7fed6b3c251666c5de4d0a78b438aeab2d1fbb80dd3f87aba261b73043", "generation_seconds_batch": 24.99434576299973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9276564121246338, "target_z": -0.322265580239897, "ce": 6.078212261199951, "genuineness": 4.494794845581055, "identity": 0.6711546778678894, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/218c40312ac1d2f607a057970ab23053e208833aaecb1e608c4f9fa1add775ad.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": "21fbc5fe0619936bf6d22a5457431749915de116b4a39433c49c9725c227c1a4", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "ba171526c57f423f4071617e51a974b5c773a1691a5c7b6258660c3f855a14d4", "generation_seconds_batch": 24.93912238604389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1503562927246094, "target_z": -0.5898439111363709, "ce": 6.0305562019348145, "genuineness": 4.7834930419921875, "identity": 0.6217337250709534, "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_TENS_reg_low/21fbc5fe0619936bf6d22a5457431749915de116b4a39433c49c9725c227c1a4.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": "2204394aec56ef0bd0edf89554a1f1417dc10733e6ebe4a11fa070404660b786", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "21981db738e312be4bc9b90269ae816ac46bdc8250c75c030631c96aafab29fc", "generation_seconds_batch": 19.85497943300288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5537810325622559, "target_z": 0.1269531160479794, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.6924002170562744, "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_TENS_reg_low/2204394aec56ef0bd0edf89554a1f1417dc10733e6ebe4a11fa070404660b786.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "2253d3a52fb888890ec6111f8b53315dea58f4384baa33152fbd78ec7b2f02ad", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8f3b92544427d53b08a8a48cfc9cb25a31505b1bd7a6bfdc6834d1155ade03ee", "generation_seconds_batch": 25.664533686009236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205624580383301, "target_z": -0.6562498567676703, "ce": 6.133810997009277, "genuineness": 4.425758361816406, "identity": 0.6023037433624268, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2253d3a52fb888890ec6111f8b53315dea58f4384baa33152fbd78ec7b2f02ad.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "225f4dfd523ba744fe8114fd0c04a39834e9f0a3ac33449af21c06f15c1dd8e8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "24647f3e44587e97a9b7e65a01fe7f100e08bc2437d99e3d68b180fb2415eef2", "generation_seconds_batch": 15.968356427038088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.636988639831543, "target_z": 0.02697752283696265, "ce": 6.324435234069824, "genuineness": 4.482243061065674, "identity": 0.6021783351898193, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/225f4dfd523ba744fe8114fd0c04a39834e9f0a3ac33449af21c06f15c1dd8e8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "226596e59342d900a8af7fb355330d68869cdde7dbeb104a4d10a3f0b9b9568b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0cf0606d16f2feea81f022509bb300de139e8870fefd97eb6dd8564786844b7", "generation_seconds_batch": 22.229551516007632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5667853355407715, "target_z": 0.11132818766414425, "ce": 6.09409761428833, "genuineness": 4.268857479095459, "identity": 0.5907268524169922, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/226596e59342d900a8af7fb355330d68869cdde7dbeb104a4d10a3f0b9b9568b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "238e7560cfda605a0ec7dd6a2d94cea4a33df9113aadcc58db55b46abc37a5e0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85166ffa89864e3b67a152a308e6e6230a2dcbc526678533532a3ba2ab08a67b", "generation_seconds_batch": 24.33025464101229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.366843342781067, "target_z": 0.35156246419191756, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.7650570869445801, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/238e7560cfda605a0ec7dd6a2d94cea4a33df9113aadcc58db55b46abc37a5e0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "2493889e486d7aac5912a84e0526a7cfd0445e499ceb46107ce1d16c17165831", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "872d8f85e09a8575e6e6f9a58b37df8ec2152c7d233a4a746122e32deb9702c7", "generation_seconds_batch": 24.79242120392155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5407766103744507, "target_z": 0.14257818766414426, "ce": 6.141753673553467, "genuineness": 4.3504462242126465, "identity": 0.19092142581939697, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2493889e486d7aac5912a84e0526a7cfd0445e499ceb46107ce1d16c17165831.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "24a97051353f996796bd5b99c87f405d6978659609b51b301c0ee22374cf08c7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ffa327202ace4230fac041db4137d3b6ce1baacb33fdc07c580d9715e0b6b84", "generation_seconds_batch": 19.84058120008558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5712556838989258, "target_z": 0.1059569752998712, "ce": 6.38003396987915, "genuineness": 4.557555198669434, "identity": 0.1928195208311081, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/24a97051353f996796bd5b99c87f405d6978659609b51b301c0ee22374cf08c7.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": "24d33eae359a10a7206330cb4c6df18f4653d8736b7e1b57fc03554e241d5deb", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fc94eaa0d2776d4d18224d349b5fa751e18326cc1abc8fa925cbd6dc3b02b57", "generation_seconds_batch": 18.909186114906333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8951455354690552, "target_z": -0.2832031160479794, "ce": 6.356205940246582, "genuineness": 4.519899368286133, "identity": 0.5198866128921509, "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_TENS_reg_low/24d33eae359a10a7206330cb4c6df18f4653d8736b7e1b57fc03554e241d5deb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "25e2652962eccf0b72fbf5d8be240b1fe834319e035fe5f89d600d2610094c9b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc4d89548fb8968f649707a5fb29e6ec98bbcebb7fb93f8316b0e94df4a384bd", "generation_seconds_batch": 22.812596139032394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.756974220275879, "target_z": -0.11718753580808243, "ce": 6.324435234069824, "genuineness": 4.331617832183838, "identity": 0.25753146409988403, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/25e2652962eccf0b72fbf5d8be240b1fe834319e035fe5f89d600d2610094c9b.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": "264230bb189aad278e17769f99891e575bef559db1ed06eccbb61c498b273a79", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0f7d73bde343f13ef2dc290f44faf5ba81981e74823b8a96d783bbd1e204c98", "generation_seconds_batch": 19.85007185896393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6439987421035767, "target_z": 0.018554745688133947, "ce": 6.165581703186035, "genuineness": 4.595211505889893, "identity": 0.6324458122253418, "hypothesis": "De sleutels liggen op de keukentafel. Keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/264230bb189aad278e17769f99891e575bef559db1ed06eccbb61c498b273a79.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "2680f9b745f0307d9b2978dc7d6718293b9d17f62466ebc3b509a076e09fdd40", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d76dc817f0245ba85967a178dd56cd305409a757756d5448978e6ed07702cc54", "generation_seconds_batch": 16.75578307500109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.642271637916565, "target_z": 0.020629895681029624, "ce": 6.1894097328186035, "genuineness": 4.419482707977295, "identity": 0.8195078372955322, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2680f9b745f0307d9b2978dc7d6718293b9d17f62466ebc3b509a076e09fdd40.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "26ce371143720aa2644dcaa8e1a7e05ccb39da0392b2580a43a3335ee4fcba63", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2b362005f36fd15aee83e89515d22afcc27230a8545680f8cd3b30f5313090e", "generation_seconds_batch": 21.058796579018235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.5429472327232361, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/26ce371143720aa2644dcaa8e1a7e05ccb39da0392b2580a43a3335ee4fcba63.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very relaxed and unstrained. 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": "a9f5cc1ceba163fa779e86680aa5c237d5d776771e906cf6ee8075b9ed517e30", "generation_seconds_batch": 18.010320142027922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.675290584564209, "target_z": -0.019043024700128794, "ce": 6.38797664642334, "genuineness": 4.375550270080566, "identity": 0.17988911271095276, "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_TENS_reg_low/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "26fd3388ac2244b43a663cf06aa07ab4b198292885ca000825947913eac19678", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "fbe5c6f92f0043c0f148fc7f6818e420873fd3810622244f2ff6dfde90164012", "generation_seconds_batch": 19.4367338080192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.616974115371704, "target_z": 0.05102537160195621, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.7153775095939636, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/26fd3388ac2244b43a663cf06aa07ab4b198292885ca000825947913eac19678.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": "27080b9b188c42b18940df8a1c79ca6adf2a3d4e1f0c732b687068a25ab26dea", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "39c625bcf4872bdbc7988a542b7e286100b2fb305952b357a50de0dac7aa68b9", "generation_seconds_batch": 18.900832889950834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4383673667907715, "target_z": 0.26562492838383517, "ce": 6.046441555023193, "genuineness": 4.632867813110352, "identity": 0.655526340007782, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/27080b9b188c42b18940df8a1c79ca6adf2a3d4e1f0c732b687068a25ab26dea.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "2816914f796fea9a0916569c9cd5a375307289724494954af339547f3761b731", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "386a3a91793e7bc1d0fab6cda87db7d598533cef04d89cf5d315d238d287224d", "generation_seconds_batch": 19.29479461198207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.474129319190979, "target_z": 0.22265623209595878, "ce": 6.419747352600098, "genuineness": 4.344170093536377, "identity": 0.3297104239463806, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2816914f796fea9a0916569c9cd5a375307289724494954af339547f3761b731.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "281b82bd9cbbbf001ce06d5977421093b740cdb894ced40592bc058ca7fb975a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92b9da568993acee7a1e8224ae9896d585d073766aaca6f3bfade35b3610f65d", "generation_seconds_batch": 23.09495379193686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9471629858016968, "target_z": -0.3457031160479794, "ce": 6.324435234069824, "genuineness": 4.024092197418213, "identity": 0.38541269302368164, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/281b82bd9cbbbf001ce06d5977421093b740cdb894ced40592bc058ca7fb975a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "28202f9bd0d5a449e03328f0e4d39413db88f25af89b36fe19323ed79eba5655", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c150a8b312c092c9022c0e4868104f48f6a1370a5939e81d58c436fab4f6c07", "generation_seconds_batch": 23.018153270008042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3700944185256958, "target_z": 0.3476562320959588, "ce": 6.356205940246582, "genuineness": 4.582659721374512, "identity": 0.45484429597854614, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/28202f9bd0d5a449e03328f0e4d39413db88f25af89b36fe19323ed79eba5655.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "2844e94f1b4c8bd4e2843fc2e4a58fb19c208760c256634df704936864773d80", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "691d38398dea2ab9335832f266567c1cfa3892794ecaf23a9bfb83a289c2355e", "generation_seconds_batch": 26.629237808985636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4107329845428467, "target_z": 0.29882818766414426, "ce": 6.078212261199951, "genuineness": 4.055472373962402, "identity": 0.6223865747451782, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2844e94f1b4c8bd4e2843fc2e4a58fb19c208760c256634df704936864773d80.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "285ae6015bb3c61649ac2857f215f97eaf1f412130b1e48823b97e0923188d9c", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "dcfc82abd1025018e6389090edfee51f52afdebcea7f8929855584de612f68d2", "generation_seconds_batch": 16.75578307500109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6519232988357544, "target_z": 0.00903323333806955, "ce": 6.1894097328186035, "genuineness": 4.545003414154053, "identity": 0.7197867035865784, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/285ae6015bb3c61649ac2857f215f97eaf1f412130b1e48823b97e0923188d9c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "28b7f1240e23e5286efe0dcbb42387bee5c2e2cd153e6282dcb4ee5e94b99dc5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47434405d63624924d6539153eba754e7164f23b9cf4c34e66fb767626e27ba6", "generation_seconds_batch": 21.510994166019373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6586287021636963, "target_z": 0.0009765580239896964, "ce": 6.268836498260498, "genuineness": 4.1998209953308105, "identity": 0.2396199107170105, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/28b7f1240e23e5286efe0dcbb42387bee5c2e2cd153e6282dcb4ee5e94b99dc5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "28bfb19e544585cf4b1999607d4bdf6972d93a79ccfbaf75db43aa1cad3900aa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a3ba9210ac14f4f4e0e1226b34bf8b79c1be4e7bd28d85fd66f19227a13c01c", "generation_seconds_batch": 24.761599438963458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4676271677017212, "target_z": 0.23046869628787636, "ce": 6.252951145172119, "genuineness": 4.21237325668335, "identity": 0.6407632231712341, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschaffen.", "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_TENS_reg_low/28bfb19e544585cf4b1999607d4bdf6972d93a79ccfbaf75db43aa1cad3900aa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "290b1269468e7c6bbbbc2e09b3cb9ae4f980e0a044c8a0f2bb95bf98b048dc30", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96b80761e82fd6c1011e0792f344aa47991d54c6eaf4b3a47e08b25ea2c116d8", "generation_seconds_batch": 24.49595363799017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.783795714378357, "target_z": -0.14941409383207213, "ce": 6.117925643920898, "genuineness": 4.570107460021973, "identity": 0.7095077037811279, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/290b1269468e7c6bbbbc2e09b3cb9ae4f980e0a044c8a0f2bb95bf98b048dc30.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "29868d0b630ad6ba22690784e9ead56fc3018fa452d527aaf27f8d3df33d04a5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6edc9dca9a9fea9147e9402f66d0da8bf93cfe7b92381c7d8554a47d1a341491", "generation_seconds_batch": 23.09495379193686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1406028270721436, "target_z": -0.5781249283838351, "ce": 6.125868320465088, "genuineness": 3.7793264389038086, "identity": 0.3628588616847992, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/29868d0b630ad6ba22690784e9ead56fc3018fa452d527aaf27f8d3df33d04a5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "29b89784e0ec1892fd3a9294b8fd3f8929a90459e815c5803d1e7f342b479846", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0065d10181c228181eae10f7e575e7f13d3863e9d99b10042a3a3ba28c3c953b", "generation_seconds_batch": 24.63003994396422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.58466637134552, "target_z": 0.08984376790404121, "ce": 6.0305562019348145, "genuineness": 4.143336772918701, "identity": 0.6608948111534119, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/29b89784e0ec1892fd3a9294b8fd3f8929a90459e815c5803d1e7f342b479846.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "29c0ee42628f0dc2446a54a6a08f4387bc2336f116acb0bfdfd782ed94af571d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbf6f0413a15b943fe046be6a7a9326a5ac3902647448281e134a9b074e780bd", "generation_seconds_batch": 26.25702152797021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5415894985198975, "target_z": 0.14160148640782483, "ce": 6.284721851348877, "genuineness": 4.532451152801514, "identity": 0.7126737236976624, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/29c0ee42628f0dc2446a54a6a08f4387bc2336f116acb0bfdfd782ed94af571d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "2a0436a3b66b80a3330942488cceeb4d77d62a12694e2f48d986d9e1120df8fb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0b949ded9f5f50b02886d0607f7409d176c81878388b0f3b3118b706ac3df3c9", "generation_seconds_batch": 17.04239165200852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.697235345840454, "target_z": -0.045410091347850595, "ce": 6.157639026641846, "genuineness": 3.9299514293670654, "identity": 0.49874934554100037, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2a0436a3b66b80a3330942488cceeb4d77d62a12694e2f48d986d9e1120df8fb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "2a1c0b6f39eed1c1271e1bd0ff1183f5319ee7ce1c4a2c90e1427a1a2ad702d3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8571e74eb4b7e18e724deb2eb443191b315b0217c89b6f96e8ea44bef832c06b", "generation_seconds_batch": 26.74374882504344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0268146991729736, "target_z": -0.4414062320959588, "ce": 5.792276382446289, "genuineness": 4.080576419830322, "identity": 0.6085933446884155, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "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_TENS_reg_low/2a1c0b6f39eed1c1271e1bd0ff1183f5319ee7ce1c4a2c90e1427a1a2ad702d3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "2a4d9298c6598a53e1215a3c31108c063ae59d9c8153c224a634d835f5c81beb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bd41ea263a048ec01d46e47e28f73a05c1fdb7782369e6fe6f5e6466a515d33", "generation_seconds_batch": 26.856478002970107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.586698293685913, "target_z": 0.08740237284406698, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.6563757658004761, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2a4d9298c6598a53e1215a3c31108c063ae59d9c8153c224a634d835f5c81beb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "2a8941835de1a9360ab5d209089078299742f39e7be4705e845a3e82f08841d0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dbf62d7a7e84a6ff88291b1ec296b1e212114be82d13909f40feff14152f9e5", "generation_seconds_batch": 25.06517980399076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5602831840515137, "target_z": 0.11914065185606182, "ce": 5.561938762664795, "genuineness": 3.9299514293670654, "identity": 0.6236839294433594, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "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_TENS_reg_low/2a8941835de1a9360ab5d209089078299742f39e7be4705e845a3e82f08841d0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "2ac01b5496f1b536f1f32f76a6409efca4c6bbbb474b14cecde0c063321b8341", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ebfa37eddca393d404f01c2cbf7163cdc9f77db7123df2993d848b4e9333b32", "generation_seconds_batch": 26.325407224008814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.615262031555176, "target_z": -1.1484373925757527, "ce": 6.109982967376709, "genuineness": 4.256305694580078, "identity": 0.07051604241132736, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2ac01b5496f1b536f1f32f76a6409efca4c6bbbb474b14cecde0c063321b8341.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": "2ad51dcfea4299ce12f83b06db95e583efc90e88ecc4d7dc9e1648e066dc8f73", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94432d5c14015c2a932cd67b5cdfaec79d7065c3a22431d0f07f4688519fae1e", "generation_seconds_batch": 17.181772356969304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.883766770362854, "target_z": -0.26953130371212364, "ce": 6.149696350097656, "genuineness": 4.6705241203308105, "identity": 0.695008397102356, "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_TENS_reg_low/2ad51dcfea4299ce12f83b06db95e583efc90e88ecc4d7dc9e1648e066dc8f73.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": "2b03f1bd8e243c1ea834e257fd00ddbd3de1e22d8595ecebb515d75a6be80ef3", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "c0b1fe5e2024906d09be6d2f4cf8f5256c4568330af6bb0de07e773df48e083b", "generation_seconds_batch": 21.017601352999918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5212701559066772, "target_z": 0.16601558023989696, "ce": 6.062326908111572, "genuineness": 4.833701133728027, "identity": 0.6936710476875305, "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_TENS_reg_low/2b03f1bd8e243c1ea834e257fd00ddbd3de1e22d8595ecebb515d75a6be80ef3.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": "2b5c641a0762d42bf0ece03abf392112769d38185ac095485fc7a5f44c56df0c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "43d09c40712fc97fba2aee3c392bb38e737e100b9c15925a5750bfe9a51a29df", "generation_seconds_batch": 20.07211304199882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9244053363800049, "target_z": -0.3183593481439382, "ce": 6.284721851348877, "genuineness": 4.545003414154053, "identity": 0.668881893157959, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2b5c641a0762d42bf0ece03abf392112769d38185ac095485fc7a5f44c56df0c.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": "2ba1272981fca906322bd07e7c6cfa7f2006cc8acfc6aae20c2b79ef29668997", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "cd583dc5eb2837ea93711aee9057387b4e87d6def865fe0ecace0ad25350a17b", "generation_seconds_batch": 25.25871634494979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7439697980880737, "target_z": -0.10156246419191757, "ce": 6.395919322967529, "genuineness": 4.344170093536377, "identity": 0.453474760055542, "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_TENS_reg_low/2ba1272981fca906322bd07e7c6cfa7f2006cc8acfc6aae20c2b79ef29668997.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "2bd724836e5124f91177066613a83a4aaeda1fa4e7fe22d53cdf5ce97c6fdb2f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fb24534378f7d21375a58e14bd4563c5352db2f067a9a45130ae2c10f89cb47f", "generation_seconds_batch": 17.36488949705381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.604172945022583, "target_z": 0.06640623209595879, "ce": 6.348263263702393, "genuineness": 4.2374773025512695, "identity": 0.5171185731887817, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2bd724836e5124f91177066613a83a4aaeda1fa4e7fe22d53cdf5ce97c6fdb2f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "2bdf40252be79ce192cc9cbe1f8b8a4dedefafa199495d744622fac69f71a4dc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "765b0b19d3a1613d956d7ef63fcc03cc227e649888fab3ba34d100fddb17f3bb", "generation_seconds_batch": 24.537829334032722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2251312732696533, "target_z": -0.6796875358080824, "ce": 6.1020402908325195, "genuineness": 4.281409740447998, "identity": 0.6376208066940308, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2bdf40252be79ce192cc9cbe1f8b8a4dedefafa199495d744622fac69f71a4dc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "2c052c601f52704fde3bcf7a80f0f98a8e3f0de263c866226e079004aa49a011", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8128c0f247152c6c96f4561e724f9b5214c8c2c9cc825738134bb9fe7d007195", "generation_seconds_batch": 18.314719072077423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.693374752998352, "target_z": -0.0407715123500644, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.3497329354286194, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2c052c601f52704fde3bcf7a80f0f98a8e3f0de263c866226e079004aa49a011.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "2c40843fa7ac6d9b7f43b510ddaacb73e529970ac2bf360f70e3e3f6881a332d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62b37e20d853e994cb791a1a1d86876fbb0202e17bb122ac79f8b47c95a09ab3", "generation_seconds_batch": 24.377054914017208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6112847328186035, "target_z": 0.05786127776988408, "ce": 6.38003396987915, "genuineness": 4.243753433227539, "identity": 0.5658815503120422, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2c40843fa7ac6d9b7f43b510ddaacb73e529970ac2bf360f70e3e3f6881a332d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "2c9d80d636b802f7674daa2288e11a4ae7b379b557c1166dff24ac2666e9a41e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4107f1dc9710a53169f1b8033008b0e067a37cf83b9434787b280012615255de", "generation_seconds_batch": 22.81185071298387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7480336427688599, "target_z": -0.10644525431186605, "ce": 6.419747352600098, "genuineness": 4.369274139404297, "identity": 0.6454808712005615, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2c9d80d636b802f7674daa2288e11a4ae7b379b557c1166dff24ac2666e9a41e.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": "2cb6803ee9b9d91cd18e52838f3c15bfa3efece2974c3becc15a1c5758305cf1", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "08baeeb94c7d73d614817ccf42b25783b31b01f479fbc0d7a0fb9eba9256ded1", "generation_seconds_batch": 24.3374675209634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.607424020767212, "target_z": 0.0625, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.6739493608474731, "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_TENS_reg_low/2cb6803ee9b9d91cd18e52838f3c15bfa3efece2974c3becc15a1c5758305cf1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "2d2d07248a1311496d3a736b8805e328bb8cac47e7c90c7062f8e327f42072c1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec48a275cfdccb10c07ec3b4b69a26d45d1846190ea6cc2be13affd86840f3e0", "generation_seconds_batch": 17.924870752030984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4026052951812744, "target_z": 0.3085937679040412, "ce": 6.165581703186035, "genuineness": 4.645420074462891, "identity": 0.5530297756195068, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2d2d07248a1311496d3a736b8805e328bb8cac47e7c90c7062f8e327f42072c1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "2d67d399d561be069f0c2ba916ebc6b144c7d3e29095a828b80f1d99b06dc9b3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd7e4121e203803bf4fdfd94b219a14280a026cb5af91a67e02b0e245724e72b", "generation_seconds_batch": 26.748798663029447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6480627059936523, "target_z": 0.01367181233585575, "ce": 5.77639102935791, "genuineness": 4.6705241203308105, "identity": 0.5872690677642822, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2d67d399d561be069f0c2ba916ebc6b144c7d3e29095a828b80f1d99b06dc9b3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "2d992c92648656c2b77d579a146457ce7607f139160c8a61a9559f85d1da0fde", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d103b9415c188ed698d8a724a5c4aca8f44624efdd1b963d82f30fff72c6878c", "generation_seconds_batch": 19.65862367209047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4123585224151611, "target_z": 0.29687507161616483, "ce": 6.356205940246582, "genuineness": 4.570107460021973, "identity": 0.3699987232685089, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2d992c92648656c2b77d579a146457ce7607f139160c8a61a9559f85d1da0fde.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "2db31eb7ec09f2195462723a6146b42d40504a48394264d14a578986f2c7bec5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ff369b6e27a1b8f2dac12e37f9a37a673f874467be4a7dab5f188a77010a9cb", "generation_seconds_batch": 23.45474422397092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3635921478271484, "target_z": 0.3554688395202061, "ce": 5.80021858215332, "genuineness": 4.149612903594971, "identity": 0.6729626655578613, "hypothesis": "la puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2db31eb7ec09f2195462723a6146b42d40504a48394264d14a578986f2c7bec5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "2dfe3f4f7b3cb11ab0f00c2d4058fb96488b7a97a012b481b0ff9e3b0fac8341", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0cd8423345e2e55f34967e6b293380f001c2b63b10e605ed67bc84dac896a7e", "generation_seconds_batch": 24.7092287420528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8805155754089355, "target_z": -0.26562492838383517, "ce": 6.348263263702393, "genuineness": 4.256305694580078, "identity": 0.4268449544906616, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2dfe3f4f7b3cb11ab0f00c2d4058fb96488b7a97a012b481b0ff9e3b0fac8341.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "2ee28cd84cfea3995e01ce2678505efbc13015ca16247b054ba1852a2479201e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "fab5e2c556654b98d02650c92c735b65f4ebdc114fcb337befa8b109dac0b27e", "generation_seconds_batch": 24.17586968606338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4611248970031738, "target_z": 0.23828130371212364, "ce": 6.0146708488464355, "genuineness": 4.262581825256348, "identity": 0.6912621855735779, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/2ee28cd84cfea3995e01ce2678505efbc13015ca16247b054ba1852a2479201e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "2f2896072610484aacf590d1d110bc20b5196643e1f9f3766cc2d7d063a0f682", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd4089732176d16ee9c3afb55c44089595005eadb120693c23192f4fe7e85a70", "generation_seconds_batch": 22.193142421077937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.57003653049469, "target_z": 0.10742181233585575, "ce": 6.181467056274414, "genuineness": 4.13078498840332, "identity": 0.2907676100730896, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2f2896072610484aacf590d1d110bc20b5196643e1f9f3766cc2d7d063a0f682.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "2f56d9697b05f5edfab09fec30a5d303925c6dd9cfdfd66a29e9f5ecf2350270", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65c044abc6ba9125a771b8254c875ae81b3e231e9ff0fa5d893535d36a950833", "generation_seconds_batch": 20.159247908974066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.130849599838257, "target_z": -0.5664062320959588, "ce": 5.982900142669678, "genuineness": 4.507347106933594, "identity": 0.6909744739532471, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/2f56d9697b05f5edfab09fec30a5d303925c6dd9cfdfd66a29e9f5ecf2350270.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "2f5bb41f231e5275e0f4747f3591450aa1bcdb333441040c4efeee56e751d555", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a110828c0ef1b4a071fb19a93c764df509d95d9e0383ccacd974abec01c65657", "generation_seconds_batch": 24.64905165194068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3863499164581299, "target_z": 0.32812492838383517, "ce": 6.0146708488464355, "genuineness": 4.400654315948486, "identity": 0.7352626323699951, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/2f5bb41f231e5275e0f4747f3591450aa1bcdb333441040c4efeee56e751d555.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": "2fa75f695cffeeffcf4376bb2d09fad2071397b452ab52011daf988e0f47e5dc", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3c3dcbf1b5737e35b48abd40169255cb4bff352a4d4567e1b2b2723e5c9a35b", "generation_seconds_batch": 26.47911919397302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8455663919448853, "target_z": -0.2236327901199485, "ce": 6.221180438995361, "genuineness": 4.143336772918701, "identity": 0.3541043996810913, "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_TENS_reg_low/2fa75f695cffeeffcf4376bb2d09fad2071397b452ab52011daf988e0f47e5dc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "30cfb442c2b5cf209389ce5b1444f5c5d9353f48c2a4c748f1e1bb7371cac672", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ff5a716c7d5af6449939690449e40f873569a949f6b531a7cd5c02cc65e90c9", "generation_seconds_batch": 21.72825229400769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3681793212890625, "target_z": -0.8515626074242473, "ce": 6.2767791748046875, "genuineness": 4.444586753845215, "identity": 0.7573212385177612, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/30cfb442c2b5cf209389ce5b1444f5c5d9353f48c2a4c748f1e1bb7371cac672.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "30ecd4441b2be80d535351efb87ca0e99577ddbb7fe40b2353c21a6c994cd186", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "281c9474c2bb76c075dd9bb8b9b4edbefde588a55b6c01694559faf5af0ad705", "generation_seconds_batch": 27.53181885892991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6200220584869385, "target_z": 0.04736320739582999, "ce": 6.3641486167907715, "genuineness": 4.243753433227539, "identity": 0.360227108001709, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/30ecd4441b2be80d535351efb87ca0e99577ddbb7fe40b2353c21a6c994cd186.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "3104e546f655df1124b86e3862b7e1a3b6a2d69f3ca6df55da4124d6c9394272", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7325762600f5376cf268fb36ac90e6de7b9625dbc44613bf34d949b24774a7e3", "generation_seconds_batch": 25.008290038094856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.52045738697052, "target_z": 0.16699213826388667, "ce": 6.213237762451172, "genuineness": 4.7834930419921875, "identity": 0.7649096846580505, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3104e546f655df1124b86e3862b7e1a3b6a2d69f3ca6df55da4124d6c9394272.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "3132312592703d8306639c22fa6f93f00fd119db551f0481168023df2079822d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "4556898daea5a972a215069ec9990bfdc443961f3425c6281ff9ade6db82855c", "generation_seconds_batch": 18.982315422967076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5830408334732056, "target_z": 0.09179688395202061, "ce": 6.507116317749023, "genuineness": 4.419482707977295, "identity": 0.6917927265167236, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3132312592703d8306639c22fa6f93f00fd119db551f0481168023df2079822d.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": "314be7ef6ed7ad919d2d1971ab076ad43f44b150530956ab05015ccbb928bd5c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73260792314be39389b2491dd7f2dd2da18ece80ddb25896d9a460becb3649f0", "generation_seconds_batch": 22.939864960033447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5627214908599854, "target_z": 0.11621097778409273, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.766207218170166, "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_TENS_reg_low/314be7ef6ed7ad919d2d1971ab076ad43f44b150530956ab05015ccbb928bd5c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "31a8ec83c502ad70368f3fab120802bdc6a778adadfd7bee1ae62fe8d5770389", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27589cac19241560f36f0ca6df465e8015952044986f4d095412358d152d0e17", "generation_seconds_batch": 26.25702152797021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.730965495109558, "target_z": -0.08593753580808243, "ce": 5.990842819213867, "genuineness": 4.432034492492676, "identity": 0.6018679141998291, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/31a8ec83c502ad70368f3fab120802bdc6a778adadfd7bee1ae62fe8d5770389.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": "31ada0e9a7f4b0e8218cf6a19bd4045ac4fe31e931ac9e12bd1a923269e9ca41", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82cd7d5b57ee119b6cdee5012593f3627a84d688d489577e85cfb5ca86952476", "generation_seconds_batch": 19.899905817001127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5789769887924194, "target_z": 0.09667967407196909, "ce": 5.95112943649292, "genuineness": 4.68307638168335, "identity": 0.7796521186828613, "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_TENS_reg_low/31ada0e9a7f4b0e8218cf6a19bd4045ac4fe31e931ac9e12bd1a923269e9ca41.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "31b0d774d0a7b1f96c73e4f220d4898b15fe119bf3d15e79a285ddc7ca053eb2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1a6a9bcb8a4fd668e37d99cf544a64d1720207b4dcfb4c9d773374fda0e6975a", "generation_seconds_batch": 18.520342689938843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6635053157806396, "target_z": -0.004882790119948482, "ce": 6.332377910614014, "genuineness": 4.595211505889893, "identity": 0.43617337942123413, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/31b0d774d0a7b1f96c73e4f220d4898b15fe119bf3d15e79a285ddc7ca053eb2.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": "3225d3557c33f761b2935471f25693033404f9da33ae179b8dc1111b40e00ef1", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69fdc9866adabd619735d523b38e1c3f5896ba2074da97aa47093f78286df180", "generation_seconds_batch": 26.06688578496687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5050146579742432, "target_z": 0.1855468839520206, "ce": 6.4515180587768555, "genuineness": 4.256305694580078, "identity": 0.4430340528488159, "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_TENS_reg_low/3225d3557c33f761b2935471f25693033404f9da33ae179b8dc1111b40e00ef1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "3236f1641bab648ec551ba1863508248009d09e1520b041a277b993c6f0e9428", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74c16d648c3e2a196cd3a876f2040c182e9b31ed7f3bbdfe2765dd79a051603e", "generation_seconds_batch": 17.03915092605166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6423732042312622, "target_z": 0.02050786173611334, "ce": 6.403861999511719, "genuineness": 4.469690799713135, "identity": 0.35739994049072266, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3236f1641bab648ec551ba1863508248009d09e1520b041a277b993c6f0e9428.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "3250de37b9d0e0a9ec3b21240078558ed4d1f4ed68f03c14a76743962d3548a0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e281301a012bbbd01510c23d18ed634e4515e5c3ded541534c9790256d4e209", "generation_seconds_batch": 23.779829851002432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.994303822517395, "target_z": -0.4023437679040412, "ce": 6.125868320465088, "genuineness": 4.557555198669434, "identity": 0.5341442823410034, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3250de37b9d0e0a9ec3b21240078558ed4d1f4ed68f03c14a76743962d3548a0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "32954afc09d77c21c8e6bb1608c912d05481ec7ab8ad88ca60c2318335e4ff43", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fe76d9db2f18170fa6087d82ca00c3b06cb7f210d698a93d5dfbd18b88acc19", "generation_seconds_batch": 27.096251641050912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6405445337295532, "target_z": 0.022705045673925298, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.7316937446594238, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/32954afc09d77c21c8e6bb1608c912d05481ec7ab8ad88ca60c2318335e4ff43.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "32a8467fbc80e9cf29f8625f3170a946e45ab7e29c16a9e67af825ec62d44783", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7667b6ca140c8107a6c89da17152ce86d3f5e46604def9bbb5399d15b15be488", "generation_seconds_batch": 24.237824577023275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.728933572769165, "target_z": -0.08349614074810818, "ce": 5.752562999725342, "genuineness": 3.9299514293670654, "identity": 0.4934438467025757, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/32a8467fbc80e9cf29f8625f3170a946e45ab7e29c16a9e67af825ec62d44783.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "32a9b52baca172c4bcc090b1232aad10f67db4fc8cf8f54145c4fab2ea52d922", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0564ab086f96549f370a2da2bea5cbcd5c984e2e278ac4125a6cb6c1fbec21b7", "generation_seconds_batch": 25.052163051092066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9536651372909546, "target_z": -0.353515580239897, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.7482904195785522, "hypothesis": "Die Tür 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_TENS_reg_low/32a9b52baca172c4bcc090b1232aad10f67db4fc8cf8f54145c4fab2ea52d922.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": "32df7fa8e2ef3dcf447a4973f39a2f9e6b1166bbbbe4bdb3c28feeb33fb56edb", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "8238ecd145ebdde34be0971e4ac2a64c3bf4c01e886e710f954bcbba144945ba", "generation_seconds_batch": 20.291013185982592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6399348974227905, "target_z": 0.02343753580808243, "ce": 5.7247633934021, "genuineness": 4.362998008728027, "identity": 0.6667124032974243, "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_TENS_reg_low/32df7fa8e2ef3dcf447a4973f39a2f9e6b1166bbbbe4bdb3c28feeb33fb56edb.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": "32f94def75bb2b40a48ecccc9a7de008c177ab2932bd9d62cfdf9dd8bb8119d5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a2208538b6c9ff748e10fe69c64cfca867cb6e6fd90a7078328935c649d24016", "generation_seconds_batch": 17.38362170290202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.639731764793396, "target_z": 0.023681603697914996, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.6389970779418945, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/32f94def75bb2b40a48ecccc9a7de008c177ab2932bd9d62cfdf9dd8bb8119d5.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": "331ced75fa6180dcd29b83581cc1c35a86184c33ed6ab35a80f6b0410dcaa1f2", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "03baa641646ef0c9c65c1079cc8ce97d910bacc5dc564c9678e38d3181befcbe", "generation_seconds_batch": 19.06373547494877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.7811489701271057, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/331ced75fa6180dcd29b83581cc1c35a86184c33ed6ab35a80f6b0410dcaa1f2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "335396ae3615b9e90f8c6438538c5c883a011a9afbdb7e74bd9e9a767da26494", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e548838f41599b44e7bf1af65a0579a9e107544676120698b6a07ce0e72ae035", "generation_seconds_batch": 25.643444009008817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6730554103851318, "target_z": -0.016357418517992273, "ce": 6.268836498260498, "genuineness": 4.293961524963379, "identity": 0.6394109725952148, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/335396ae3615b9e90f8c6438538c5c883a011a9afbdb7e74bd9e9a767da26494.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "33bff59a257cf138f67b4055b072af17882527e77da7eb5409c9b09a6719e3be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "101476c792cc137287df89f11a233183e2d319fc723812d2b4b770863cd384b4", "generation_seconds_batch": 24.212607741937973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.626930594444275, "target_z": 0.03906246419191757, "ce": 6.300607204437256, "genuineness": 4.557555198669434, "identity": 0.5094467401504517, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/33bff59a257cf138f67b4055b072af17882527e77da7eb5409c9b09a6719e3be.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very relaxed and unstrained. 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": "6a89863d6b8b7138956730d94378ab712ff0fe8e1257a2fef3a12507bd6c079f", "generation_seconds_batch": 17.4102937389398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": -0.644531160479794, "ce": 6.260893821716309, "genuineness": 4.545003414154053, "identity": 0.8055238723754883, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.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": "3400f47ed4d7fb41040611a4c9702b39879dae75b191a50acc5e49df2f2eaf34", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "312f5dfba77601c7bdd8501422f3e4e52f9728eb4231f9c903a83b59f4b8ff06", "generation_seconds_batch": 18.74141849996522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5318361520767212, "target_z": 0.1533203259280309, "ce": 5.871703147888184, "genuineness": 4.65797233581543, "identity": 0.687888503074646, "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_TENS_reg_low/3400f47ed4d7fb41040611a4c9702b39879dae75b191a50acc5e49df2f2eaf34.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "34235c105b27e08d37fd134f688b26e91b778627b36950550301231ac5a8701d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "4fca93d02212b5f375d0d90cb40194f74ecac7a8c28b0ef60aae2f09fbd223de", "generation_seconds_batch": 16.53578114800621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.883766770362854, "target_z": -0.26953130371212364, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.5850721001625061, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/34235c105b27e08d37fd134f688b26e91b778627b36950550301231ac5a8701d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "3434603ce8d3d78e8e7b845ab2e2a8313037046ae5e754c80c08261538d4be99", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9cfb36162302e49335e3cf1227b546480b68bec130927e37dece9d3289c6d4bb", "generation_seconds_batch": 26.687958552036434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4253629446029663, "target_z": 0.28125, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.5725798606872559, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3434603ce8d3d78e8e7b845ab2e2a8313037046ae5e754c80c08261538d4be99.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "347934e1799409b54ce685820c8f81396a29c17ba1586b650de6661d75659177", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7c6dc3c259d3d15f9af63fdb0b1233896b3a8aae2b7024361262bbd89f9205e1", "generation_seconds_batch": 24.70449912897311, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5948259830474854, "target_z": 0.07763679260417, "ce": 6.403861999511719, "genuineness": 4.18726921081543, "identity": 0.5036139488220215, "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_TENS_reg_low/347934e1799409b54ce685820c8f81396a29c17ba1586b650de6661d75659177.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very relaxed and unstrained. 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": "a7c0d34db34279bb6c999c6069549af921611c8905ced7321ecff73db08cef8d", "generation_seconds_batch": 21.618182509904727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9455374479293823, "target_z": -0.34375, "ce": 6.4515180587768555, "genuineness": 3.9927117824554443, "identity": 0.5203720927238464, "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_TENS_reg_low/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.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": "3590ea4cad3744ea267645703bb8f299562d4e507e99f532df547615a1b9428e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51abcbdcae4f9850729092a759c8c5249f4cc2279edd2f5e7ba07c568912c5f8", "generation_seconds_batch": 27.74315555789508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3896009922027588, "target_z": 0.32421869628787636, "ce": 6.292664527893066, "genuineness": 4.494794845581055, "identity": 0.4601403474807739, "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_TENS_reg_low/3590ea4cad3744ea267645703bb8f299562d4e507e99f532df547615a1b9428e.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": "35d89b686945d474568fc140c37d233a4effb234f3f756967d815c07155fd551", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3639560ae28386e7b3b5349a43be61f15bd3349ea238f226e9e39f5065a62c7", "generation_seconds_batch": 20.279402674990706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.780544638633728, "target_z": -0.14550786173611333, "ce": 5.784333229064941, "genuineness": 4.7081804275512695, "identity": 0.7322582602500916, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, il est bien, et vérifie les deux derniers chiffres. Appelle-moi quand tu auras 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_TENS_reg_low/35d89b686945d474568fc140c37d233a4effb234f3f756967d815c07155fd551.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "364aff9fd4a66f375021b068500880a2e8b3b4ea4b4ba289cab8fd3a74d94a31", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e305d2044075b2f177a45ebad94d7ae03dd0d04a7c3f19004704d6840b52385", "generation_seconds_batch": 23.389638917986304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5253340005874634, "target_z": 0.1611327901199485, "ce": 6.268836498260498, "genuineness": 4.607763767242432, "identity": 0.7691453099250793, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/364aff9fd4a66f375021b068500880a2e8b3b4ea4b4ba289cab8fd3a74d94a31.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "36b24320d87380396d62f99a4456d9255aba29b164ab10f938b166af5e425b1e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d69795a6fb85f3e5d9c9a3840864077fd7073d0942a4b233e95867581f966be2", "generation_seconds_batch": 18.178208232973702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3681793212890625, "target_z": -0.8515626074242473, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.7009050846099854, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/36b24320d87380396d62f99a4456d9255aba29b164ab10f938b166af5e425b1e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "36ec1e9984005904ae117a3591fa969a36b488c87021537577d097eebc43eb7a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "423f2ec2c183c7d2fe7351f62914b1221bd739ac3952b0c226c2e4d6954672c7", "generation_seconds_batch": 17.778710516053252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.774855136871338, "target_z": -0.13867181233585574, "ce": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.7598479986190796, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/36ec1e9984005904ae117a3591fa969a36b488c87021537577d097eebc43eb7a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "373a9bc60c37da61586ac0ad0a6aaf04c12d76db96a3551a0fab2d11146a585f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48303b41bf3f3ede54226b3c65144ea23d7cc717c6086107136291f93a453240", "generation_seconds_batch": 17.286489492980763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0268146991729736, "target_z": -0.4414062320959588, "ce": 6.086154937744141, "genuineness": 4.337893962860107, "identity": 0.7271285057067871, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/373a9bc60c37da61586ac0ad0a6aaf04c12d76db96a3551a0fab2d11146a585f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "37c4339f505cd2873a8725521a8fc2f121b824f76288b55942592c5396a3b033", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56f063f45d06ad70749c2dcc4962a44691f9a0d35731295814bf2f1a75834b39", "generation_seconds_batch": 25.35504110401962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6277433633804321, "target_z": 0.038085906167927876, "ce": 6.284721851348877, "genuineness": 4.105680465698242, "identity": 0.3571104407310486, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/37c4339f505cd2873a8725521a8fc2f121b824f76288b55942592c5396a3b033.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": "37dc38276d6dd94da7eee1307241db429de1c1ae43e202428414403b86f8c69e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a7380eee71f6890eab91ec8df93d4c040e70052aa263ccb0dc5884b716e815f", "generation_seconds_batch": 24.06995008594822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3961031436920166, "target_z": 0.3164062320959588, "ce": 5.895530700683594, "genuineness": 4.720732688903809, "identity": 0.5256561636924744, "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_TENS_reg_low/37dc38276d6dd94da7eee1307241db429de1c1ae43e202428414403b86f8c69e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "37febed66aeb1b944a4d91ca4bc57cac9d527ebc7c379d1e638a3becd837ad6d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2c16fde9bce1f0534c2dbd78e827ddf45d20953c1df3b61856bb3c1842301f1", "generation_seconds_batch": 23.42114908702206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4497461318969727, "target_z": 0.2519531160479794, "ce": 5.6413655281066895, "genuineness": 4.13078498840332, "identity": 0.5426295399665833, "hypothesis": "la puerta sigue abierta he dejado tu abrigo sobre la silla dime cuándo estás listo", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/37febed66aeb1b944a4d91ca4bc57cac9d527ebc7c379d1e638a3becd837ad6d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "3829aa9c968110d7f31a48473b298282dd5f3268192becdd3e233d2e078d5f0b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac2de99212141dbd0ddd7401625e17357e370401025dec358e07d1a4fa70799d", "generation_seconds_batch": 24.701112549984828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.80330228805542, "target_z": -0.17285162964015455, "ce": 6.348263263702393, "genuineness": 4.2249250411987305, "identity": 0.7089016437530518, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3829aa9c968110d7f31a48473b298282dd5f3268192becdd3e233d2e078d5f0b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "3852f5899720b4d7dd2e6bf6cce6d80127d64b47c858a80e83064e0e6d10ca05", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "703f2e68f50a0f5664857ce931f5ff07e898129706bf64876cd0d84dd7e3fc6c", "generation_seconds_batch": 17.174523120978847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9471629858016968, "target_z": -0.3457031160479794, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.3596688508987427, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3852f5899720b4d7dd2e6bf6cce6d80127d64b47c858a80e83064e0e6d10ca05.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "38fdcfdff6f064b5dde973798feae84b3ab771fa157a67102c32b54b82d83639", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a13f828b47a43c888bb6acde4905deaa713d74d038e60b4587d391704d306bce", "generation_seconds_batch": 25.15223058196716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361966609954834, "target_z": 0.35742195556818546, "ce": 5.752562999725342, "genuineness": 4.394378662109375, "identity": 0.6899975538253784, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "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_TENS_reg_low/38fdcfdff6f064b5dde973798feae84b3ab771fa157a67102c32b54b82d83639.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "396adfba29f425241a4727c91f363917b5b0ea65a5616534999422547a74bcb4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d39ccad30e270073a52990311885d59fe6d4da35ef51a6c4decd592a8ebe7bad", "generation_seconds_batch": 24.761599438963458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6545648574829102, "target_z": 0.0058593481439381786, "ce": 6.213237762451172, "genuineness": 4.2374773025512695, "identity": 0.5211823582649231, "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_TENS_reg_low/396adfba29f425241a4727c91f363917b5b0ea65a5616534999422547a74bcb4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "3a32bcbb18996c9b081cd0d5f7278a96fc5fd1250eb627df8b2f5b9f200baa5c", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "abde6c181f259cc1929117cbe41739d1f334690cd0c5e238c9a8f2954e8b80a0", "generation_seconds_batch": 17.819917223067023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4332010746002197, "target_z": -0.9296875358080824, "ce": 6.197352409362793, "genuineness": 4.256305694580078, "identity": 0.7677171230316162, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3a32bcbb18996c9b081cd0d5f7278a96fc5fd1250eb627df8b2f5b9f200baa5c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "3a681bb480e35a74918288334a025b219e9c51c8bbdafd89369def8eaa537854", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cfccae446f7953996cd9b4fe95b4d362004a9078b5d50e51c49a586f75ade49", "generation_seconds_batch": 26.362437876989134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7382804155349731, "target_z": -0.0947265580239897, "ce": 5.808161735534668, "genuineness": 4.193545341491699, "identity": 0.5890811681747437, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3a681bb480e35a74918288334a025b219e9c51c8bbdafd89369def8eaa537854.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "3a99e1fb013e57665126591d3c2096221cd87fcf69b48e69ce8755e41ad2f264", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "4e77243c234f045ec14246db3dcf9cd06245bb45dc851ba125f644e757c8ce71", "generation_seconds_batch": 16.325509530026466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9260308742523193, "target_z": -0.32031246419191756, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.6699820756912231, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3a99e1fb013e57665126591d3c2096221cd87fcf69b48e69ce8755e41ad2f264.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "3ab891133724a2fedc9f42a4a7498997eb9dae4d30106a184a3f4bb58526b322", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "3218885e2a9d182a43844a4aaba0180b5cd513b9e75bd277e9df7c7f5f9d9ac5", "generation_seconds_batch": 16.20615777000785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7862340211868286, "target_z": -0.15234376790404122, "ce": 6.1020402908325195, "genuineness": 3.9927117824554443, "identity": 0.23396870493888855, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3ab891133724a2fedc9f42a4a7498997eb9dae4d30106a184a3f4bb58526b322.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "3b1e42cbdedab75300cb38cd77a08fe8897b058d643cf9ee548df036973ef3bb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "edb41f2cf54ad8119e0748f9f6f2ff0d2ad4bed39979754e788f695b1fc0834a", "generation_seconds_batch": 16.24275882903021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1015896797180176, "target_z": -0.5312498567676703, "ce": 6.141753673553467, "genuineness": 3.7416701316833496, "identity": 0.29330626130104065, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3b1e42cbdedab75300cb38cd77a08fe8897b058d643cf9ee548df036973ef3bb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "3b3bb1148213b9d9545952c90ea74ab5c2eb7d6b7258346e80b28497c7589534", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8c78b825b04d26a9ad3337f85ef9b53843a82f022296543e765ea7cf13f2422", "generation_seconds_batch": 16.541026238002814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.596858024597168, "target_z": 0.07519525431186605, "ce": 6.435632705688477, "genuineness": 4.3127899169921875, "identity": 0.6425394415855408, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3b3bb1148213b9d9545952c90ea74ab5c2eb7d6b7258346e80b28497c7589534.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": "3b49408f3729e172b1f0a04c2430539c03e077318606148194ba624064fba0f3", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "91d1c9c99b02859811b0cfcf61dc7b4a26ad787c5e2f6d7e2873b34442e9fd55", "generation_seconds_batch": 20.627625997993164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6960161924362183, "target_z": -0.04394525431186606, "ce": 5.998785495758057, "genuineness": 4.7081804275512695, "identity": 0.7308239340782166, "hypothesis": "J'écris le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand il y aura un moment.", "wer": 0.22727272727272727, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3b49408f3729e172b1f0a04c2430539c03e077318606148194ba624064fba0f3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "3bbdcaaed9ace9d23e7218787982459b4225a8ac5c14548e260e47817ea0a841", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db4861ec90976483786a7b9e62cbf388e6d7e7b7d59a18f66573090635194d5e", "generation_seconds_batch": 24.79854503890965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.579383373260498, "target_z": 0.09619139505997425, "ce": 5.998785495758057, "genuineness": 4.20609712600708, "identity": 0.5385730266571045, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3bbdcaaed9ace9d23e7218787982459b4225a8ac5c14548e260e47817ea0a841.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "3bcbd9d606437268e4ae2e1e800e9141e0d8356e49f5415bb0e7a13019817465", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46a2eaa5ff877306c72f2005ed7317de40a02b642b3a9e12c0a2bfaed1123976", "generation_seconds_batch": 23.636896810028702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5732876062393188, "target_z": 0.10351558023989696, "ce": 5.665193557739258, "genuineness": 4.105680465698242, "identity": 0.6172629594802856, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3bcbd9d606437268e4ae2e1e800e9141e0d8356e49f5415bb0e7a13019817465.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "3bffd0db562d3a31f64cfed4460c44445d8dc3505ec24b118663463456f6caf7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34aefea14d7b48902c809d282fff806d2dde1503cb7d9ae5d5557225088f7bce", "generation_seconds_batch": 24.75148405705113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4302395582199097, "target_z": 0.2753906518560618, "ce": 6.443574905395508, "genuineness": 4.482243061065674, "identity": 0.470517098903656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3bffd0db562d3a31f64cfed4460c44445d8dc3505ec24b118663463456f6caf7.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very relaxed and unstrained. 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": "9c8de8e8b4fa35064dec2987f4213c0305d226876c9395e28039ce7fca26c270", "generation_seconds_batch": 18.284540520980954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.7809360027313232, "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_TENS_reg_low/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "3caa735c0e1ff8b378d23d0168c128b2ee079d7ea99291ac5998062d832690a0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8439ab959eb5a4b8e070939e356e3ccc0cc2f3a30ce42d1dfac0e6f4f1ce4c5", "generation_seconds_batch": 25.008290038094856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3489623069763184, "target_z": 0.3730468839520206, "ce": 6.213237762451172, "genuineness": 4.482243061065674, "identity": 0.7365429401397705, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3caa735c0e1ff8b378d23d0168c128b2ee079d7ea99291ac5998062d832690a0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "3cee829d34c492ee95cf648bd4f1f2609bb2cdb27537d8fb37baec1331689935", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15131c79573a066f78a8e20087ad0aedfa5b88b461b4edeaa2aaf14d9c79e95d", "generation_seconds_batch": 24.77653999393806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7082078456878662, "target_z": -0.058593767904041215, "ce": 5.518253803253174, "genuineness": 4.099404335021973, "identity": 0.6759846210479736, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3cee829d34c492ee95cf648bd4f1f2609bb2cdb27537d8fb37baec1331689935.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "3cf1bb45831b9eb950fe93c38d7afd0e141943f3ee2181c050b4e81df01f56de", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "588eea8d1f78c6c1c7b9f478f34510e02fa6ab2a7f29bbdbaf085a8d2c3b6cf5", "generation_seconds_batch": 24.367105231038295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4960741996765137, "target_z": 0.19628902221590727, "ce": 5.95112943649292, "genuineness": 4.545003414154053, "identity": 0.48038631677627563, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3cf1bb45831b9eb950fe93c38d7afd0e141943f3ee2181c050b4e81df01f56de.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": "3d2672817ef6c6d0b35b0b4639b740c7994786cd44ff10d9f9a6b5eb978b29c5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f782f7579b62fbd9db798d7efbd38e869cbddd79f9fcfbdbedcaede6a1764c16", "generation_seconds_batch": 19.85497943300288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.807366132736206, "target_z": -0.17773441976010304, "ce": 6.197352409362793, "genuineness": 4.632867813110352, "identity": 0.631939172744751, "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_TENS_reg_low/3d2672817ef6c6d0b35b0b4639b740c7994786cd44ff10d9f9a6b5eb978b29c5.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": "3d37f2bac5a77a2a04ce457ac0ebc316f17135ddb5dd368259162b59622af115", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "161cfb9f20413a736e3053cb41d099b5a933f9804c824ec383e7f4718b288f12", "generation_seconds_batch": 25.011490056989715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3603410720825195, "target_z": 0.35937507161616483, "ce": 6.0305562019348145, "genuineness": 4.532451152801514, "identity": 0.5110511183738708, "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_TENS_reg_low/3d37f2bac5a77a2a04ce457ac0ebc316f17135ddb5dd368259162b59622af115.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "3d3fb19aa3336d1558d1e30af892d03cd0764408a64f6783181c0122472f9cc1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "943788810fc1051d932f3d925f0fbf3d98a96d4cd47979f739fd13977440d215", "generation_seconds_batch": 21.72825229400769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1080920696258545, "target_z": -0.5390626074242473, "ce": 6.141753673553467, "genuineness": 4.287685871124268, "identity": 0.701678991317749, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3d3fb19aa3336d1558d1e30af892d03cd0764408a64f6783181c0122472f9cc1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "3dbde4d6915b90825888b8326d19ec4aac2f8e0fa8189d1e148a968f78b6a509", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6cfba11ee78156262639bb52e40a8d2b2f5cd5dc247fa55957193618763f385b", "generation_seconds_batch": 24.265701323049143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6334327459335327, "target_z": 0.03125, "ce": 6.252951145172119, "genuineness": 4.545003414154053, "identity": 0.4272230863571167, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3dbde4d6915b90825888b8326d19ec4aac2f8e0fa8189d1e148a968f78b6a509.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": "3dda11bd615397d7aebb301ae55f2a9d0b73e99a550b546987b47cf56659e125", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d81ade40de90287f381c4a6ef6a646b875009592887b5f00728ee2de4d21658", "generation_seconds_batch": 18.891724584042095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5212701559066772, "target_z": 0.16601558023989696, "ce": 6.046441555023193, "genuineness": 4.833701133728027, "identity": 0.7673210501670837, "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_TENS_reg_low/3dda11bd615397d7aebb301ae55f2a9d0b73e99a550b546987b47cf56659e125.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "3de74d7215ce792af16b1d799057e20c327d2d72267481eced32ced184e32df8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ca848633faec5aac2e842333626bf14b82ce8c506c002a6c90e0e0fd56d4d3f5", "generation_seconds_batch": 19.349533853004687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6759001016616821, "target_z": -0.01977537160195621, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.6339991092681885, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3de74d7215ce792af16b1d799057e20c327d2d72267481eced32ced184e32df8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "3df05b12b06fb4286f690e0e886616f941660865be8e0bc67b9ec9e49b0d4481", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4dbe49023b689e7cd72d8489413b4f29d707395f35dd9734fbe8e91b31508f6a", "generation_seconds_batch": 19.84058120008558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5448405742645264, "target_z": 0.13769525431186605, "ce": 6.284721851348877, "genuineness": 4.413206577301025, "identity": 0.4000939130783081, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3df05b12b06fb4286f690e0e886616f941660865be8e0bc67b9ec9e49b0d4481.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "3e601c8df3b99d0c973ebbed66307fcedfcd58263d47113f04bf7b66a6c32ade", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c9b92d800db9d64954e82bdd1d393a60c0b2107e1b897c4ead2f6e17b6e13522", "generation_seconds_batch": 24.595487658050843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7114589214324951, "target_z": -0.0625, "ce": 6.308549880981445, "genuineness": 4.519899368286133, "identity": 0.6512608528137207, "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_TENS_reg_low/3e601c8df3b99d0c973ebbed66307fcedfcd58263d47113f04bf7b66a6c32ade.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": "3e775a9081cf30687f1a7aec5cce643321f8b5c820b330d39ee29405a9fc2e8a", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "baa3b5c0ba724e90559c76e73d6d21e380cef850b4be1a1e0581bcd0d7765101", "generation_seconds_batch": 26.19616402790416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2969448566436768, "target_z": 0.4355468839520206, "ce": 5.959072113037109, "genuineness": 4.494794845581055, "identity": 0.7369730472564697, "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_TENS_reg_low/3e775a9081cf30687f1a7aec5cce643321f8b5c820b330d39ee29405a9fc2e8a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "3ed3ec54db6cfd3440d2e646ebb92d85f4c77582f442a85d8a2d6d6ec6d42b24", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e74d1a62930fb5e4f2486f764832e8050a88465b7c21806129e096477cf957a1", "generation_seconds_batch": 23.949956815107726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.994303822517395, "target_z": -0.4023437679040412, "ce": 6.133810997009277, "genuineness": 4.3504462242126465, "identity": 0.6102403998374939, "hypothesis": "Ik 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_TENS_reg_low/3ed3ec54db6cfd3440d2e646ebb92d85f4c77582f442a85d8a2d6d6ec6d42b24.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "3edd78dc015866793aefaf7d7cb3f527c92eae28b798c1f06c7ed7e9cfe92c62", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff22612dff872a8a08fd110f2c791c6b9adc0184c3d64cdbc15b4ac601d518f0", "generation_seconds_batch": 16.074291587923653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5017635822296143, "target_z": 0.1894531160479794, "ce": 6.229123115539551, "genuineness": 4.570107460021973, "identity": 0.8180785179138184, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3edd78dc015866793aefaf7d7cb3f527c92eae28b798c1f06c7ed7e9cfe92c62.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "3eea7a17a6f2c3100429bf74f6567426d5ecf6fcacf163342fd347fe0439097d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e3e822dc8bdcffd562ee2242ad3620ca53a90080ce46e4f5a3618d433acd253", "generation_seconds_batch": 24.212607741937973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8244342803955078, "target_z": -0.19824213826388667, "ce": 5.935244083404541, "genuineness": 4.444586753845215, "identity": 0.6352511644363403, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3eea7a17a6f2c3100429bf74f6567426d5ecf6fcacf163342fd347fe0439097d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "3eff69eb520d37cc45ad44635d26c6010be160bad664055e807c05a8b88c2258", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "bc17040036562bbf26537a8e7303f8d3d7862e579704eb648703af78a2190f05", "generation_seconds_batch": 24.77653999393806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3132003545761108, "target_z": 0.416015580239897, "ce": 5.764476776123047, "genuineness": 4.268857479095459, "identity": 0.6478286385536194, "hypothesis": "Tu puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás linda.", "wer": 0.13333333333333333, "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_TENS_reg_low/3eff69eb520d37cc45ad44635d26c6010be160bad664055e807c05a8b88c2258.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "3f215f7edfa0e676f54e352e12f3b8b7df7635f6c55b648e7c6207cd50a23802", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcecbe16bc57901d11a8a90d95bfd9846b53fef96824b3043dd5af17928ca1e1", "generation_seconds_batch": 27.730835248017684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1717778444290161, "target_z": 0.5859375358080824, "ce": 5.816103935241699, "genuineness": 4.494794845581055, "identity": 0.4728885889053345, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3f215f7edfa0e676f54e352e12f3b8b7df7635f6c55b648e7c6207cd50a23802.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "3f48b8cab9ff1353111c58efbb21eb50018f495fcc86878717940d410c27c19b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c6492cc51519ecdb5558c2f607c2efd870cf317cb11710ba0076b52b12cc8e7", "generation_seconds_batch": 23.81093512394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.682605504989624, "target_z": -0.027832046916036062, "ce": 6.054384231567383, "genuineness": 4.231201171875, "identity": 0.7063872814178467, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen haar samen in.", "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_TENS_reg_low/3f48b8cab9ff1353111c58efbb21eb50018f495fcc86878717940d410c27c19b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "3fefbd4d0aa11aa23943b1bbe1eda379425afe1e9a9d2b0cd192cf899e03a108", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5a5ff0e4cd417b8b511c8f53d68d1a71457b392e0ee8227721f92b7f43bd8da", "generation_seconds_batch": 23.917896409053355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498222827911377, "target_z": -1.0078124641919175, "ce": 6.157639026641846, "genuineness": 4.356721878051758, "identity": 0.6042696237564087, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/3fefbd4d0aa11aa23943b1bbe1eda379425afe1e9a9d2b0cd192cf899e03a108.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "3ff6e0aa6c15ef5200fdd8f8a147b89fb940004c99d117605e6e24d2d6cb3817", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7787977a89d72c5c7ae688958f20c2d05648373e0a869dac1bf26a613a5b8a01", "generation_seconds_batch": 23.486624783021398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.397728681564331, "target_z": 0.3144531160479794, "ce": 6.157639026641846, "genuineness": 4.7332844734191895, "identity": 0.7639080882072449, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/3ff6e0aa6c15ef5200fdd8f8a147b89fb940004c99d117605e6e24d2d6cb3817.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": "402bb3f3b10e610392a0d83681c69eab74a9c7bcec8e4360da7d3cadcb9edd55", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee4e9f798bb4c403a0e76c013795c194ae76c3a555afe0c2ab22cc5ecee28d39", "generation_seconds_batch": 24.16592906008009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9634184837341309, "target_z": -0.365234419760103, "ce": 5.95112943649292, "genuineness": 4.3881025314331055, "identity": 0.6582318544387817, "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_TENS_reg_low/402bb3f3b10e610392a0d83681c69eab74a9c7bcec8e4360da7d3cadcb9edd55.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "4054c0c9383471d2c692b5a2cfe105ef233a56b6765260f014cf68b291d0bcc1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "b390d86fc5ff21daadfaeddf1cc23d388053759d9b669016c82ebec55b819cfb", "generation_seconds_batch": 25.65446751692798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3538389205932617, "target_z": 0.36718753580808244, "ce": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.6331884860992432, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinaus.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4054c0c9383471d2c692b5a2cfe105ef233a56b6765260f014cf68b291d0bcc1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "405d99ad956e739e35108c66b646c0edd24843858e8f9eecea08f62bb0405ce9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "711eee275c42831c11a12561ef46e7e22e282f454dffcdcf8a1af4fb8cbbd337", "generation_seconds_batch": 22.601890597026795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9455374479293823, "target_z": -0.34375, "ce": 6.23706579208374, "genuineness": 4.425758361816406, "identity": 0.6327164173126221, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/405d99ad956e739e35108c66b646c0edd24843858e8f9eecea08f62bb0405ce9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "406271a4f00e252a53cd87526e9cb30dc0edc8ea2864d5a2d237908305cb86be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08dc15ddfa11b4d535d1d513627cc11c5ae7a0985a6bca8630d21c3524980407", "generation_seconds_batch": 24.7092287420528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8431280851364136, "target_z": -0.2207031160479794, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.189230278134346, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/406271a4f00e252a53cd87526e9cb30dc0edc8ea2864d5a2d237908305cb86be.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "40bcb177a5bde09d618afc0949c662f16dae3860858027798dbcf806d98d1b11", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be21c78d84c921e2bbb980ad5af6cd1bd87e7d99f76e2b4094ab7158a02bb7a0", "generation_seconds_batch": 24.199689535889775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4806314706802368, "target_z": 0.21484376790404122, "ce": 5.935244083404541, "genuineness": 4.444586753845215, "identity": 0.5491083860397339, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "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_TENS_reg_low/40bcb177a5bde09d618afc0949c662f16dae3860858027798dbcf806d98d1b11.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "40c992b6ed60de476710eeef2f02279615fce7c601a3c9a60785c1ce2254a0bd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e084c8e214de0ca9cabe04df626b7dafefd3b03afb8208d1798e079c8b34ae1", "generation_seconds_batch": 24.507337045972236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7667274475097656, "target_z": -0.12890623209595878, "ce": 6.38003396987915, "genuineness": 4.507347106933594, "identity": 0.5350969433784485, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/40c992b6ed60de476710eeef2f02279615fce7c601a3c9a60785c1ce2254a0bd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "40ff5c07f88389c138c9d2977a5c799f6c45675ed30e553ad25025d0bd14ae95", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2b6b04d88f406a84589392c39f2980ff00e505359294a2e43b6fd99afcbaa584", "generation_seconds_batch": 23.100803704932332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7935489416122437, "target_z": -0.1611327901199485, "ce": 6.332377910614014, "genuineness": 4.250029563903809, "identity": 0.6266301274299622, "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_TENS_reg_low/40ff5c07f88389c138c9d2977a5c799f6c45675ed30e553ad25025d0bd14ae95.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "4113557ae10ecab73e87f06e5a528a133bcef3d5f5095353f7eb397663aa8119", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "994636c36f3990594e15ae0e906e0b1b2a596cf301c9eec730daa1a34db438d9", "generation_seconds_batch": 15.866210230975412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5789769887924194, "target_z": 0.09667967407196909, "ce": 6.507116317749023, "genuineness": 4.218649387359619, "identity": 0.7087842226028442, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4113557ae10ecab73e87f06e5a528a133bcef3d5f5095353f7eb397663aa8119.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "41967e24802919017f1b6816efe75d22eb16172369c66d5b0dc2fe1c505f880c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "789c566aa53026b091d3f5be0a58dfc19ca8936f92006367b960f8781eaf5248", "generation_seconds_batch": 22.493943006033078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6253050565719604, "target_z": 0.041015580239896966, "ce": 5.557967185974121, "genuineness": 4.262581825256348, "identity": 0.6797418594360352, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar junto", "wer": 0.29411764705882354, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/41967e24802919017f1b6816efe75d22eb16172369c66d5b0dc2fe1c505f880c.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": "4221b9c339ef5648929892919312c9aac5853d148780be86cee316efdf6c02af", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9ba59252899ddc3f732243355caeb7ffb323fafcd6e829b8597e55f997ae2fbd", "generation_seconds_batch": 20.19667174795177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3356683254241943, "target_z": -0.8125, "ce": 6.173524379730225, "genuineness": 4.532451152801514, "identity": 0.41861075162887573, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4221b9c339ef5648929892919312c9aac5853d148780be86cee316efdf6c02af.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "425282583d8470cd4eb89abb256221a91cac56f5a92a742ed571fe411a12a325", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "54b54bc9aba8c1f2a5d408c5cdd9d385a480a42e72b1e340bf293d59694957a3", "generation_seconds_batch": 18.946965656941757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8886433839797974, "target_z": -0.2753906518560618, "ce": 6.229123115539551, "genuineness": 4.362998008728027, "identity": 0.43840137124061584, "hypothesis": "I found the Little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/425282583d8470cd4eb89abb256221a91cac56f5a92a742ed571fe411a12a325.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "42b11abca779db78840e389b63faebc60a1bfbc5ca0735aa192b25ad272c04bd", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a29aef5ace7038e2daead7c1889b6c43c40fbc76775f1065ad15e79282013651", "generation_seconds_batch": 16.53960423404351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.277148723602295, "target_z": -0.7421875358080824, "ce": 5.879645347595215, "genuineness": 4.325342178344727, "identity": 0.735664963722229, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/42b11abca779db78840e389b63faebc60a1bfbc5ca0735aa192b25ad272c04bd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "42c30e951e105a95e35b3d5c1ce2e61a84d31425ac3a45b5ac91381074c00a7e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "391b4683408a9d554bc59aca05b0444435242f367e865752c6f7dea06c464afb", "generation_seconds_batch": 23.76468881999608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6251018047332764, "target_z": 0.04125979136205925, "ce": 6.403861999511719, "genuineness": 4.099404335021973, "identity": 0.49678972363471985, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/42c30e951e105a95e35b3d5c1ce2e61a84d31425ac3a45b5ac91381074c00a7e.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very relaxed and unstrained. 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": "2d4b8a104396d9a46c96600e63a9e645ab83c0c2a874cbf5e6198b7f8dea540d", "generation_seconds_batch": 17.4102937389398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2934041023254395, "target_z": -0.7617186962878764, "ce": 6.046441555023193, "genuineness": 4.457139015197754, "identity": 0.5262365341186523, "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_TENS_reg_low/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "43c72f66237a30d82c71af85f42b0ada55d7ab35d8905a851138bd6474154805", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a166b3b4ab2029bdcf4b21d1901ae81b8c2c791c9f562e168e39eb7892462e6a", "generation_seconds_batch": 18.672299705911428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7870467901229858, "target_z": -0.1533203259280309, "ce": 6.117925643920898, "genuineness": 4.532451152801514, "identity": 0.4107034206390381, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/43c72f66237a30d82c71af85f42b0ada55d7ab35d8905a851138bd6474154805.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very relaxed and unstrained. 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": "de97671f58974cbd541bc17837dde513481b74f96c5e7618d6d14907eb28fa1f", "generation_seconds_batch": 20.559019336942583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8561325073242188, "target_z": -0.23632818766414426, "ce": 6.1894097328186035, "genuineness": 4.444586753845215, "identity": 0.6616686582565308, "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_TENS_reg_low/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "441e175e1e2609537ae669521a611c09df62210f491143777ae5115fe9652698", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21cc6c623811ca3767d0c54ba9ed1ad3e06559eada33cada89f0c95a33bcb4f3", "generation_seconds_batch": 17.96617010398768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1373517513275146, "target_z": -0.5742186962878764, "ce": 6.2767791748046875, "genuineness": 4.570107460021973, "identity": 0.7498824000358582, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/441e175e1e2609537ae669521a611c09df62210f491143777ae5115fe9652698.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "4436f4456f241b2ef0e7f4b6ca6bcabc062c2c26012a20b99b20da6b5e2a59c7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79691845931cd70fb7d171bc3326e3754025d78fca7297f9a493780e06a27b88", "generation_seconds_batch": 26.8028804929927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.366843342781067, "target_z": 0.35156246419191756, "ce": 6.41180419921875, "genuineness": 4.507347106933594, "identity": 0.4555206298828125, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4436f4456f241b2ef0e7f4b6ca6bcabc062c2c26012a20b99b20da6b5e2a59c7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "44751ff0d56cddc57240cbb98027b96cb0d55425f858007e33c6e7a0bad900c7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0bd31e481042852bd0974e17ec1de2f7902c031972b6f0e63c624e74693582c5", "generation_seconds_batch": 19.687512636068277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5594704151153564, "target_z": 0.12011720988005152, "ce": 6.260893821716309, "genuineness": 4.720732688903809, "identity": 0.7288435101509094, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/44751ff0d56cddc57240cbb98027b96cb0d55425f858007e33c6e7a0bad900c7.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": "449e7a02096d268bd6e0f33ac997162155676d4945e9394b1454cf888604ef06", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14b3197619336c05d1b134cc256faca2a5bab2d0b33736a7d6484a283714754c", "generation_seconds_batch": 24.404016270069405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1080920696258545, "target_z": -0.5390626074242473, "ce": 5.990842819213867, "genuineness": 4.444586753845215, "identity": 0.5713377594947815, "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_TENS_reg_low/449e7a02096d268bd6e0f33ac997162155676d4945e9394b1454cf888604ef06.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "44b7c955de74e6f98dab97fda3b72b5ca606034b0e236cdc489810440f3aeef7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9e0096e3c4f5d1c884ae47fb5553c1c4b2e87af414f700173e567326c5c3d71", "generation_seconds_batch": 21.851931801997125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5033891201019287, "target_z": 0.1875, "ce": 6.0305562019348145, "genuineness": 4.049196243286133, "identity": 0.7088993787765503, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/44b7c955de74e6f98dab97fda3b72b5ca606034b0e236cdc489810440f3aeef7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "4591593c98e928c31d34efbb2dae4e2099b2a497d9a39492da9162b787366019", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c270470c5e26402fba8312f209b7b26c508161f04ef033e998c49aca0799d200", "generation_seconds_batch": 23.067874257918447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6823006868362427, "target_z": -0.0274658018489575, "ce": 6.419747352600098, "genuineness": 4.256305694580078, "identity": 0.678345799446106, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4591593c98e928c31d34efbb2dae4e2099b2a497d9a39492da9162b787366019.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": "45b1355811ee6f350c163d4ef20e2d69a415edd6ed9ae8e453414f3fa69e4da6", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "379f087aded1caec6715960e66c6c9e09642e172520d8224501e2778f3758769", "generation_seconds_batch": 26.19616402790416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.575319528579712, "target_z": 0.10107418517992273, "ce": 5.895530700683594, "genuineness": 4.65797233581543, "identity": 0.5117175579071045, "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_TENS_reg_low/45b1355811ee6f350c163d4ef20e2d69a415edd6ed9ae8e453414f3fa69e4da6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "45c42aec82d530d2186c5312b54416def0caa0eebdb0e308c017f73a119b5143", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b94807647e7c31958a01af3e565182b7f28f708e6ed33e01d4b42d60090838b4", "generation_seconds_batch": 24.093465792015195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8350003957748413, "target_z": -0.21093753580808242, "ce": 5.863759994506836, "genuineness": 4.281409740447998, "identity": 0.7721749544143677, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/45c42aec82d530d2186c5312b54416def0caa0eebdb0e308c017f73a119b5143.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very relaxed and unstrained. 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": "d0f758e14d180842d3c48d9e64d2019b5630d513e38016caf3a84ff9288d8986", "generation_seconds_batch": 17.862279142951593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8268725872039795, "target_z": -0.20117181233585574, "ce": 6.260893821716309, "genuineness": 4.174716949462891, "identity": 0.5793142318725586, "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_TENS_reg_low/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "461c1694810dd7b63b9a19ab28e84c3b2e2a5f44839b5d008631c40178d47f85", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "6d02de8c0c13d91afa14dc4314101f5e652f3e0c216e800d40366f3bf311a585", "generation_seconds_batch": 18.494608669891022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.241386651992798, "target_z": -0.6992186962878764, "ce": 6.141753673553467, "genuineness": 4.306513786315918, "identity": 0.6865657567977905, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/461c1694810dd7b63b9a19ab28e84c3b2e2a5f44839b5d008631c40178d47f85.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "461f08f739d8045120c9bd346e47572e06ddf7074a7750ef3a9b08ffdfd4d494", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4115893b59cea135faf9838a5cc17847060bae762ec4c1fcc969de8282f010a", "generation_seconds_batch": 23.113238853984512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6927651166915894, "target_z": -0.040039022215907265, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.33986425399780273, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/461f08f739d8045120c9bd346e47572e06ddf7074a7750ef3a9b08ffdfd4d494.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "46c6c80caf3aeadc3391d5cf77e20a5e737dbb05431b6e2ef5b4cef442a11e13", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "dc8d9ada981062d989be86de5ca028cd28e1e4b9081008bd6627fa36fd00c6cc", "generation_seconds_batch": 17.734317956026644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.70414400100708, "target_z": -0.053710977784092735, "ce": 5.903473377227783, "genuineness": 3.973883628845215, "identity": 0.6487179398536682, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/46c6c80caf3aeadc3391d5cf77e20a5e737dbb05431b6e2ef5b4cef442a11e13.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "470cacb610e8755595cc57e986309a3e10573c87e4499cf4089fa255b6e2e071", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db83126ea0befa2d0c84fa2906959b3fade5609fa7eb1542f3c0f8fa61491855", "generation_seconds_batch": 17.762931396020576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4529972076416016, "target_z": 0.2480468839520206, "ce": 6.197352409362793, "genuineness": 4.482243061065674, "identity": 0.6835383772850037, "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_TENS_reg_low/470cacb610e8755595cc57e986309a3e10573c87e4499cf4089fa255b6e2e071.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "473851db3a4bc82e8389fede72e00a2971953f7a78e76bfed7352df090d6ca33", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ed51c4b5ea837273e61b175d6d2e59ec3917a74be56e0342fbba61710dd9e18f", "generation_seconds_batch": 17.038432899978943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6897172927856445, "target_z": -0.03637700124211076, "ce": 6.0146708488464355, "genuineness": 3.9927117824554443, "identity": 0.5860235691070557, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/473851db3a4bc82e8389fede72e00a2971953f7a78e76bfed7352df090d6ca33.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "4783a6ac6790e7289bbc06fd0e31234c46e66ecb5e2754ded1d6af83776eaee2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1b61655b08fc7a6a8a5870db2be6f02edfa6b60f090ac028017416a10a8a89c1", "generation_seconds_batch": 19.063538464950398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.844753623008728, "target_z": -0.22265623209595878, "ce": 6.292664527893066, "genuineness": 4.118232727050781, "identity": 0.31217077374458313, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4783a6ac6790e7289bbc06fd0e31234c46e66ecb5e2754ded1d6af83776eaee2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "47a0f14f40a9404c623c772c1808be8b03bd9df80b645234bb3023642dc778ad", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0de0fac93c791317e7bd8dd2af8dff70ddf0b5406a6294b632cb35046c7cdb87", "generation_seconds_batch": 16.53578114800621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0154359340667725, "target_z": -0.427734419760103, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.5398035049438477, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/47a0f14f40a9404c623c772c1808be8b03bd9df80b645234bb3023642dc778ad.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very relaxed and unstrained. 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": "14a57a3daa33d43213235a3e78c62fcff967c0aeb688ae22aa89e8cdd128c215", "generation_seconds_batch": 18.647080836002715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6675691604614258, "target_z": -0.009765580239896965, "ce": 6.181467056274414, "genuineness": 4.632867813110352, "identity": 0.6744471788406372, "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_TENS_reg_low/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "47d9d9a8b7da54bcc2c4974402f3c5b9618f350846dc00d4889b317397652f5f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ab37890c7d415673522fe39684b97221fc9592277bb6cd945cf28fa1c6d68b5", "generation_seconds_batch": 24.863072050036862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511980056762695, "target_z": -0.4707032592803091, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.6031071543693542, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/47d9d9a8b7da54bcc2c4974402f3c5b9618f350846dc00d4889b317397652f5f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "48230d0786568d96ad21c7108c83ec90f970558dd5be709a8f93f4fcdd16617e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4f967de11f0a5aedca758766d4d785a08bbb733347fcf495ced2e102e62d845", "generation_seconds_batch": 23.282547803013586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7707912921905518, "target_z": -0.13378902221590727, "ce": 5.959072113037109, "genuineness": 4.758388519287109, "identity": 0.6714035868644714, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/48230d0786568d96ad21c7108c83ec90f970558dd5be709a8f93f4fcdd16617e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "4825d1ffd01313407f7d4d163ec2e1e995c6c1d615081349c2cff5a7e6559d3a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "4d6d19298132b3b66d4689b0d0b0857cf39eff83a77764a2dc5ba5208b489413", "generation_seconds_batch": 17.13349960197229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5814152956008911, "target_z": 0.09375, "ce": 6.157639026641846, "genuineness": 3.9927117824554443, "identity": 0.6763556003570557, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4825d1ffd01313407f7d4d163ec2e1e995c6c1d615081349c2cff5a7e6559d3a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "4829df3bcd815bd516f32ab3c9ffdf549499b78276de7dbd9ab0c29f6d3fd5f1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da311464e35741b786c2d508defd3c4448c9609f43c7256cdaaa84268cbd3626", "generation_seconds_batch": 26.291318524978124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8723878860473633, "target_z": -0.2558593481439382, "ce": 5.982900142669678, "genuineness": 3.873466968536377, "identity": 0.25795313715934753, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4829df3bcd815bd516f32ab3c9ffdf549499b78276de7dbd9ab0c29f6d3fd5f1.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": "4877122dbb502ac41f2d31d00edabb999043dcbc63c1b4531b62eb9ad1d28eb6", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84f233ee5c6672477445197dee3614d67f9c462bc478c663e061562b483a0fa7", "generation_seconds_batch": 20.527365706977434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9162776470184326, "target_z": -0.3085937679040412, "ce": 6.109982967376709, "genuineness": 4.796044826507568, "identity": 0.6686563491821289, "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_TENS_reg_low/4877122dbb502ac41f2d31d00edabb999043dcbc63c1b4531b62eb9ad1d28eb6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "4900a1e0c7ff76dacf598c0a369e7a58d984fd2c10b10b0d2c4035cdd0866935", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed14b4f2d85a92218f756a369043d047c5ef641637df780bfa349257be915b0c", "generation_seconds_batch": 22.897065024939366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6574095487594604, "target_z": 0.002441395059974241, "ce": 6.23706579208374, "genuineness": 4.231201171875, "identity": 0.49610984325408936, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4900a1e0c7ff76dacf598c0a369e7a58d984fd2c10b10b0d2c4035cdd0866935.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "494ebce040168415cceb02d37223100f23d45d80b4a5cfa9f29c57daacf40147", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9b55a92112b90e4b602c1c3bce5d8808a34afd3f318014327e6e66fdbe1d1f3c", "generation_seconds_batch": 19.349533853004687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7732295989990234, "target_z": -0.13671869628787636, "ce": 6.006728172302246, "genuineness": 3.9801597595214844, "identity": 0.5481944680213928, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/494ebce040168415cceb02d37223100f23d45d80b4a5cfa9f29c57daacf40147.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "497d404a74f5039247d599652b193500ecdcffca0c18c40b41ef2d78cbf66722", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d869c74b5827eae541424ab93b6812ebeb302ddb494b1ec2c01ed56b437836f", "generation_seconds_batch": 23.92976638104301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1861181259155273, "target_z": -0.6328124641919176, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.5156701803207397, "hypothesis": "Die deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/497d404a74f5039247d599652b193500ecdcffca0c18c40b41ef2d78cbf66722.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very relaxed and unstrained. 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": "2fad9fc6f342cd0267f8f75f64294f7d12d1770c88fadb1fd11827280f4dae8a", "generation_seconds_batch": 17.145708944066428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8691368103027344, "target_z": -0.2519531160479794, "ce": 6.165581703186035, "genuineness": 4.494794845581055, "identity": 0.5468032360076904, "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_TENS_reg_low/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "49da67b1337ecbc963b9ae5021059beacd6123a676cd1c9730cbf4462ebcfe0c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73108720cfce4753e043ca983a64ee14ec91a835c4720ee9f73b5a8813995843", "generation_seconds_batch": 25.375774536048993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8853923082351685, "target_z": -0.271484419760103, "ce": 6.070269584655762, "genuineness": 4.444586753845215, "identity": 0.6575548648834229, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar was.", "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_TENS_reg_low/49da67b1337ecbc963b9ae5021059beacd6123a676cd1c9730cbf4462ebcfe0c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "49dfd8c1d5b46f5034be23aafcb8958efa85790c0ae59cdb7e7067ab0176cafe", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6a0543161f0c86611f4c43273ed4b7773d64755d9ae75480ea650bf6e3dc9ce", "generation_seconds_batch": 26.06577475904487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4830697774887085, "target_z": 0.21191409383207213, "ce": 5.486483573913574, "genuineness": 4.2249250411987305, "identity": 0.569278359413147, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/49dfd8c1d5b46f5034be23aafcb8958efa85790c0ae59cdb7e7067ab0176cafe.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "4a1bff83182e0dca3c5c533267346c96025a0923ead80611a306ec707cc5c29c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 126, "seconds_generated": 10.08, "mp3_sha256": "eb7bfd89ccd2c2ee002c4c3f2e82b39c1887da9b35e723662d38fbc040c71965", "generation_seconds_batch": 18.29483273893129, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.236799716949463, "target_z": 0.5078124641919176, "ce": 5.450741291046143, "genuineness": 4.519899368286133, "identity": 0.7000167965888977, "hypothesis": "He encontrado una cajita, estaba junto a la ventana vieja, podemos mirar juntos.", "wer": 0.29411764705882354, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4a1bff83182e0dca3c5c533267346c96025a0923ead80611a306ec707cc5c29c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "4a99092d7d69c33c4fb449b92f0ba3559f4f46466746eb1468969a67bf0e1149", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1c8e8c290249559545b1d5e2ba2a18df933a023328b2c5ed2c1a63b85ea19cf0", "generation_seconds_batch": 24.642908500041813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4733165502548218, "target_z": 0.2236327901199485, "ce": 6.0305562019348145, "genuineness": 4.331617832183838, "identity": 0.8648116588592529, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4a99092d7d69c33c4fb449b92f0ba3559f4f46466746eb1468969a67bf0e1149.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "4b2be4246759c074b0a1defb725309f78f8256c8c35f855b7f4475369875f4b3", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "50aa550fd3ac3848727b5de95936d36a4b0d6a882dcdf4dc5654d06674d94619", "generation_seconds_batch": 17.778710516053252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.748846411705017, "target_z": -0.10742181233585575, "ce": 6.1894097328186035, "genuineness": 4.319066047668457, "identity": 0.546083390712738, "hypothesis": "La porte est à la ton fadzmir. La porte est encore ouverte. 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_TENS_reg_low/4b2be4246759c074b0a1defb725309f78f8256c8c35f855b7f4475369875f4b3.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": "4bafeff880af991771a90ccbdb1314835de8a131a511f7842aaebcdf1bea7ff8", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e38578364c9f27a48bb363032f2e145bdb68a3e342052d8bd1964d4fae6708f", "generation_seconds_batch": 26.47911919397302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5789769887924194, "target_z": 0.09667967407196909, "ce": 6.252951145172119, "genuineness": 4.231201171875, "identity": 0.26245182752609253, "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_TENS_reg_low/4bafeff880af991771a90ccbdb1314835de8a131a511f7842aaebcdf1bea7ff8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "4bcceea54b2d7264df264a88aa9d084e4c65eea350d46f3f30e6972bca1202de", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ecfbd6915d86ec1cfde9edb72645883337727511849e443941d1f77110fd22bb", "generation_seconds_batch": 18.880745266913436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4448695182800293, "target_z": 0.25781246419191756, "ce": 5.967014789581299, "genuineness": 4.369274139404297, "identity": 0.47724682092666626, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4bcceea54b2d7264df264a88aa9d084e4c65eea350d46f3f30e6972bca1202de.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "4c4811b1a22aed329348a623b04760cb5a53a2d5086e59db4005f9e46c838475", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "268ab66e7aed79cbbbd251adf0c0844175f207cfac7e700855cfc4f20dd3f8ca", "generation_seconds_batch": 26.947682947968133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4318650960922241, "target_z": 0.27343753580808244, "ce": 5.8875885009765625, "genuineness": 4.425758361816406, "identity": 0.6996506452560425, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4c4811b1a22aed329348a623b04760cb5a53a2d5086e59db4005f9e46c838475.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "4c9461cecd24d1fdfb3a0bda31951146e523c4f079a6841bbb88b5983d18c608", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "36fe3a8ae39780e4473c9efa5aebba434c7e9ff063f8438d3acf7acc5c768404", "generation_seconds_batch": 16.20615777000785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7329974174499512, "target_z": -0.08837893086805666, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.5350939035415649, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4c9461cecd24d1fdfb3a0bda31951146e523c4f079a6841bbb88b5983d18c608.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "4c9633ddcd9f88f9ba1de0bfadd09c3e50a93b805af777cbd8f2f8a5e8ae8c88", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6b5269229ad0efa9da0ca6df9986134e4e812e150b756f1c460ddbb4b8706b6", "generation_seconds_batch": 25.782550566014834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.082083225250244, "target_z": -0.5078124641919176, "ce": 6.006728172302246, "genuineness": 4.256305694580078, "identity": 0.44411301612854004, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4c9633ddcd9f88f9ba1de0bfadd09c3e50a93b805af777cbd8f2f8a5e8ae8c88.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": "4d983e972ccc273f35def92a8c184738e14afa6d4d06b39f76545620ffc673cb", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f9b9d9f766750fae877c3b3fa22faee61a46cab8b16090e8f9d8f853ac9b251e", "generation_seconds_batch": 18.839654837967828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6862629652023315, "target_z": -0.032226558023989695, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.6307483315467834, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4d983e972ccc273f35def92a8c184738e14afa6d4d06b39f76545620ffc673cb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "4dfa284fe28c41b18ecdda334e6503975e372d828db54b78db68a1ee46a807a4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd9d4014d7d04560e46bf0a684e0b0d7fec239e834a8a72ae698067321721307", "generation_seconds_batch": 23.436586808995344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7894850969314575, "target_z": -0.15625, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.34550708532333374, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4dfa284fe28c41b18ecdda334e6503975e372d828db54b78db68a1ee46a807a4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "4e1a13da3835e4420dbdb65f5dd3a7d1b615c1a7b6f5d71b57d88c6a2d374500", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef63cea81277f779a1ae57494207bafcbbae8b65526f27029dbe07a4cf6ab983", "generation_seconds_batch": 23.074417750001885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5489044189453125, "target_z": 0.13281246419191758, "ce": 6.022613525390625, "genuineness": 4.582659721374512, "identity": 0.6060951948165894, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4e1a13da3835e4420dbdb65f5dd3a7d1b615c1a7b6f5d71b57d88c6a2d374500.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "4e9e1a46ee8662d8b0aaaaa5c84f4a4e7534cdba73f6c5f89555f8a3e5b89ae6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "abcc7f655e3ae1efef86d0ccea8f6b40c70ce3d81152351645f57d629ec46f3b", "generation_seconds_batch": 23.332554060965776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6112847328186035, "target_z": 0.05786127776988408, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.8041658997535706, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/4e9e1a46ee8662d8b0aaaaa5c84f4a4e7534cdba73f6c5f89555f8a3e5b89ae6.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": "4f082526ddb376d907a819beb1f32a9359ad26e12ade687b9419e72619714451", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "eba2df8624d40321ee629210b96192b832c54a2a9fa59c26c1cdddcdbe2fd247", "generation_seconds_batch": 18.63830425508786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4026052951812744, "target_z": 0.3085937679040412, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.7288203239440918, "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_TENS_reg_low/4f082526ddb376d907a819beb1f32a9359ad26e12ade687b9419e72619714451.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "4f2d5df434791a80c7af2ce42141b9c155e5a8a1a876c6a7b7340bc2c434aa52", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36e09ac026f4babc42a1dea5d0bfa98c4de559764494627ce0cf5def90778ea7", "generation_seconds_batch": 24.11973971198313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9552907943725586, "target_z": -0.3554688395202061, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.6707586050033569, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4f2d5df434791a80c7af2ce42141b9c155e5a8a1a876c6a7b7340bc2c434aa52.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "4f66c67436154a7ae9f920fd2183ba19ca1574fc15bd12a5b646cfe3747ec071", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51014314785da53a3a5b067897471da2dbc5d4391ab9abe8992f0242a121d88a", "generation_seconds_batch": 24.256843673996627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.210790991783142, "target_z": 0.5390624641919176, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.24723052978515625, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4f66c67436154a7ae9f920fd2183ba19ca1574fc15bd12a5b646cfe3747ec071.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "4f8176e6b436e5c0f91ba1fab1fab63555b94765ebb3007d77cdbcd11ab2154e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aea181380e169e323301fad2b8e89f98aa9e68834b32e03ead492cd8b9f4da7a", "generation_seconds_batch": 17.762931396020576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4716910123825073, "target_z": 0.22558590616792787, "ce": 6.324435234069824, "genuineness": 4.375550270080566, "identity": 0.7482911944389343, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfahren.", "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_TENS_reg_low/4f8176e6b436e5c0f91ba1fab1fab63555b94765ebb3007d77cdbcd11ab2154e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "4fb2bcaa04f6735be97c9720b0957925ed0aae1aa4e5b0470435b748e62555e0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81c46b9a9c59bc8a7559505d822beb59eb07e7d4fd3061f4a769ca632e1d3bff", "generation_seconds_batch": 17.174523120978847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9455374479293823, "target_z": -0.34375, "ce": 6.252951145172119, "genuineness": 4.381826400756836, "identity": 0.3931776285171509, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/4fb2bcaa04f6735be97c9720b0957925ed0aae1aa4e5b0470435b748e62555e0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "4fff3fcf0699ba99af223f90ddbe578bb6ebfec4323a0583514ca4072f3d6cbc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9e972059bbcb5f028e81356d41f12e2838f5a8486bc8073f92dfe8d8ee74ad01", "generation_seconds_batch": 27.096251641050912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6769161224365234, "target_z": -0.020996140748108186, "ce": 6.332377910614014, "genuineness": 4.331617832183838, "identity": 0.728632926940918, "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_TENS_reg_low/4fff3fcf0699ba99af223f90ddbe578bb6ebfec4323a0583514ca4072f3d6cbc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "5001aed47cb185ddd145a51b93aaaf8d2ced372241b26cb07229664cfe3008be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95934f7a6f3abdd6aced55c8b93a4b4c4f5e00dd213e96fb60a99eaa8559ff83", "generation_seconds_batch": 22.649227416026406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6887012720108032, "target_z": -0.035156232095958785, "ce": 5.990842819213867, "genuineness": 4.507347106933594, "identity": 0.6993317604064941, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5001aed47cb185ddd145a51b93aaaf8d2ced372241b26cb07229664cfe3008be.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": "502a814ad46359b84233ab39ce6777d9f952ce70afb54f86b2f88613ad1e22fc", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1288ffef802677840a1acabbbb6b7137ee9b1aa2cc1fa0de7f3d81b4e22b6a2", "generation_seconds_batch": 19.48033198202029, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6960161924362183, "target_z": -0.04394525431186606, "ce": 5.732706069946289, "genuineness": 4.432034492492676, "identity": 0.7290812730789185, "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_TENS_reg_low/502a814ad46359b84233ab39ce6777d9f952ce70afb54f86b2f88613ad1e22fc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "50d3be0cba43660ca91e9712bc9d6144cee5aba64a7ac6f9c08319c6e9eadb5e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c85496aa8d58cd3fcdc9a14784a3fbca85d557f8c328d961b75366186890ecfe", "generation_seconds_batch": 24.933715575025417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6895140409469604, "target_z": -0.03613279011994848, "ce": 5.673136234283447, "genuineness": 4.036643981933594, "identity": 0.5683014988899231, "hypothesis": "Er staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/50d3be0cba43660ca91e9712bc9d6144cee5aba64a7ac6f9c08319c6e9eadb5e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "5122998e0279a57f60d51d43b53a9c485700fb9dd1e0710f1eecd292695750f2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9e9008e9f2cc237c6c3b36ccc2431d2e5333fcdc15bf19243c1ee4f192bacc2", "generation_seconds_batch": 26.236899528070353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4887592792510986, "target_z": 0.20507804443181454, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.781369686126709, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5122998e0279a57f60d51d43b53a9c485700fb9dd1e0710f1eecd292695750f2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "516279f09a51298c6021a142b487a0c01085d0142eedf8032ab5a48702dbaaa9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd50b1f6398c93b712b0adcb26c66143a48d63efe83222c0c9a6742f62c6ab67", "generation_seconds_batch": 25.347713648923673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8049278259277344, "target_z": -0.17480474568813395, "ce": 6.308549880981445, "genuineness": 4.017816066741943, "identity": 0.4668234586715698, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/516279f09a51298c6021a142b487a0c01085d0142eedf8032ab5a48702dbaaa9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "516ee1af04929e11b904c36f317951d5c91a431d6e53d3ce0cc47cd04d61b81b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a497b4c83d7f5f37c45ba6cdb5d7eb89f9e67751401a0d03ca396d4ca8fa3f10", "generation_seconds_batch": 17.703527727047913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.231633424758911, "target_z": -0.6875, "ce": 6.070269584655762, "genuineness": 4.413206577301025, "identity": 0.7018973231315613, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/516ee1af04929e11b904c36f317951d5c91a431d6e53d3ce0cc47cd04d61b81b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "51818ead2975fccdc1a5a077d401d598695cfb6fa39c49b8457377cd7052d672", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a0d749fc21c927124eb0e0199be54a049d0dbcc0f84e748e032c01de832f311", "generation_seconds_batch": 24.126441917032935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.508265733718872, "target_z": 0.18164065185606182, "ce": 6.022613525390625, "genuineness": 4.055472373962402, "identity": 0.6003544330596924, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/51818ead2975fccdc1a5a077d401d598695cfb6fa39c49b8457377cd7052d672.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "51fefc155ce357cde351a68198ef0561496b0c0407b35c4488ed5f622b200053", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "249365ad60a4ba2138455fab54e57598991181d0697b36d959db7fc378e82798", "generation_seconds_batch": 22.81185071298387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.593606948852539, "target_z": 0.07910148640782484, "ce": 6.3641486167907715, "genuineness": 4.394378662109375, "identity": 0.5907276272773743, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/51fefc155ce357cde351a68198ef0561496b0c0407b35c4488ed5f622b200053.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "524da7eb867c0cc453a6545f0b46cc8d8813db999f06f4a3947f16bcdb1381f3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e6e38a470a16cb0a5ed13453ac6287859a155b2597f495161925fd5b108b93de", "generation_seconds_batch": 18.036964202998206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2660595178604126, "target_z": 0.4726562320959588, "ce": 5.462655067443848, "genuineness": 4.21237325668335, "identity": 0.548335075378418, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/524da7eb867c0cc453a6545f0b46cc8d8813db999f06f4a3947f16bcdb1381f3.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": "534e331262a275bee5eecda520b8ef94a9baa3966df7ca028591495f3e3622bf", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b983d6fbf3d64dc4000960bfe113021187c2aeb07c2da1ba16c4b4e22672bfd", "generation_seconds_batch": 26.67380578396842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5594704151153564, "target_z": 0.12011720988005152, "ce": 6.284721851348877, "genuineness": 4.557555198669434, "identity": 0.24474366009235382, "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_TENS_reg_low/534e331262a275bee5eecda520b8ef94a9baa3966df7ca028591495f3e3622bf.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": "53b8559b83abff59f5345ad33ee9108c5d2c5929c34c287a69fa14fa079352a9", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "5160b362e28c8ab504366be18cd00db3c763516c072a443c150b8baf48e9dafd", "generation_seconds_batch": 20.717880478012376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5797897577285767, "target_z": 0.09570311604797939, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.6665037870407104, "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_TENS_reg_low/53b8559b83abff59f5345ad33ee9108c5d2c5929c34c287a69fa14fa079352a9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "53c80334a8d264db531b7e6da6b8eb0f4b45222eaae1de9e0ac711a9090dcdca", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "28fdb6159006348054d55109b1794d765983f0508011fb7acd0964220efd5b61", "generation_seconds_batch": 17.996784696006216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3912265300750732, "target_z": 0.322265580239897, "ce": 5.804190158843994, "genuineness": 4.344170093536377, "identity": 0.752070426940918, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/53c80334a8d264db531b7e6da6b8eb0f4b45222eaae1de9e0ac711a9090dcdca.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "53db6bcebdf15c3e3f100932214e9ab7851ab6ad2ca9d7dff48817663eea5e2d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d47ddb9bb6c54a5f975a9f9e923bac95fe1bfd148bb3dbfd623e6d2c14246f4", "generation_seconds_batch": 23.289168171933852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070704460144043, "target_z": -0.4941406518560618, "ce": 6.252951145172119, "genuineness": 4.137060642242432, "identity": 0.38617002964019775, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/53db6bcebdf15c3e3f100932214e9ab7851ab6ad2ca9d7dff48817663eea5e2d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "5423e23fa441c5ffc0d00ad2bdf9ba690984a3a60700db7c514b73e309bf4f3d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67c23d17d50681ea46975d8215c855c9693d159cdf44d863feab36d319769160", "generation_seconds_batch": 26.06577475904487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4237374067306519, "target_z": 0.2832031160479794, "ce": 6.0146708488464355, "genuineness": 4.557555198669434, "identity": 0.5086320638656616, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5423e23fa441c5ffc0d00ad2bdf9ba690984a3a60700db7c514b73e309bf4f3d.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very relaxed and unstrained. 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": "6bacc1806f2f6fc4b020448e8f8f059c132cf894b4492208b070d09c374dd71a", "generation_seconds_batch": 17.862279142951593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.825247049331665, "target_z": -0.19921869628787636, "ce": 6.308549880981445, "genuineness": 4.017816066741943, "identity": 0.5015518665313721, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "546cd07d58926435e5b75671fc2352de8f91918655224ae7d976a0a598d993ea", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f94cd16e2e2cb5f2fb33269116676423902f060eacb1cd8211cb012c769fd2df", "generation_seconds_batch": 18.125881711021066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5448405742645264, "target_z": 0.13769525431186605, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.5704014301300049, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/546cd07d58926435e5b75671fc2352de8f91918655224ae7d976a0a598d993ea.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "54821725074b705271a953a9cdf51f9453ec9956bbfcf69246d0aa36b9913a09", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29607a95667c4c76a8ab984ea59abcab0417ff697239b85203730aad1a4c774e", "generation_seconds_batch": 17.36488949705381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5212701559066772, "target_z": 0.16601558023989696, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.6194045543670654, "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_TENS_reg_low/54821725074b705271a953a9cdf51f9453ec9956bbfcf69246d0aa36b9913a09.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "549188fb8edb86f9bb6e7a13ce2843f20e56937e2b84968e3b8de863a44811be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a9111e851cc6c2ff55822eb3bee76c50258163775e631af46378decb555c5d31", "generation_seconds_batch": 17.45856888103299, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8463791608810425, "target_z": -0.22460934814393818, "ce": 6.173524379730225, "genuineness": 4.1998209953308105, "identity": 0.7293009757995605, "hypothesis": "Ik 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_TENS_reg_low/549188fb8edb86f9bb6e7a13ce2843f20e56937e2b84968e3b8de863a44811be.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "5539a14564a48e8ec851b850721b96fd0cb84fdc4b511d5d21e48d5c05ab9059", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "278fb1363f69855a2393d7e493a0ac68e94181c1ec8cd174ee68b88e82ea7098", "generation_seconds_batch": 19.092876061913557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4432439804077148, "target_z": 0.259765580239897, "ce": 6.078212261199951, "genuineness": 4.632867813110352, "identity": 0.71587073802948, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5539a14564a48e8ec851b850721b96fd0cb84fdc4b511d5d21e48d5c05ab9059.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "55752e57fb9d45397ea8c962b0c7ff1e873cf03b5d69d30eaf024fbfbe9e0ce5", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5e87310098b1fd8a0b100c770ef877800c1deaa0726f2dd73449a21c49cb754f", "generation_seconds_batch": 15.804846744984388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0398190021514893, "target_z": -0.4570311604797939, "ce": 6.252951145172119, "genuineness": 4.344170093536377, "identity": 0.651385486125946, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/55752e57fb9d45397ea8c962b0c7ff1e873cf03b5d69d30eaf024fbfbe9e0ce5.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": "561f7eeda41ce00170667930a97510b6223da0420926494c3bbefa95b8101d53", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "342766ec7103e735a0fbfef17c93784664d88068afdf755e23abad1b25742d7a", "generation_seconds_batch": 16.086366798030213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6895140409469604, "target_z": -0.03613279011994848, "ce": 5.903473377227783, "genuineness": 4.218649387359619, "identity": 0.7951570749282837, "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": 0.120000000000001, "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_TENS_reg_low/561f7eeda41ce00170667930a97510b6223da0420926494c3bbefa95b8101d53.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": "569b46293a3e07d16cc754a8d062cbb8c81befeb97e7912d96b760d30eeaab62", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0ecbf13f13e45227a3cf84a290899b6efd4a7bc1d4ac38aacff890544a1a5fa", "generation_seconds_batch": 19.49101643799804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8122427463531494, "target_z": -0.18359376790404122, "ce": 6.086154937744141, "genuineness": 4.432034492492676, "identity": 0.44189271330833435, "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_TENS_reg_low/569b46293a3e07d16cc754a8d062cbb8c81befeb97e7912d96b760d30eeaab62.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "56da479647c452b13f74721bd0a469f907a97893568981b556e3aed2180083b4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3d7331f916f720cf29e015f280f41b1850194eaf93678696fd6832f80d14abc9", "generation_seconds_batch": 17.631897851941176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7667274475097656, "target_z": -0.12890623209595878, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.6172653436660767, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/56da479647c452b13f74721bd0a469f907a97893568981b556e3aed2180083b4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "56eb11af1a3796769a7bc24f027c2829fa284bceb4d9637fce586bc209ce8f83", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1cbded9229d1c885188ea5d9d6dcb67a806ec95d6b8749eb9324d6e9e3f822a", "generation_seconds_batch": 25.429093181039207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5827512741088867, "target_z": -1.1093750716161648, "ce": 6.09409761428833, "genuineness": 4.319066047668457, "identity": 0.7899999618530273, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/56eb11af1a3796769a7bc24f027c2829fa284bceb4d9637fce586bc209ce8f83.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "56fae597be2791c45cb25d5f8b1528240eca1b7538b6c8acd2939c7809536740", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cf46d323b4069898e63d607679d38010a29bee00cfe0f772c40fe0ffd0d60e8b", "generation_seconds_batch": 26.69907436799258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6033601760864258, "target_z": 0.06738279011994848, "ce": 6.141753673553467, "genuineness": 4.306513786315918, "identity": 0.7489748597145081, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/56fae597be2791c45cb25d5f8b1528240eca1b7538b6c8acd2939c7809536740.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": "5786c25f105a0405ffbf86deb04795e15ea1ffbcf1000f91bd12a8fdb2b166a7", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "b37d7206e0e04480249dbf10d2c23ee8196762ad42dad3a9e11e6b8031fcc766", "generation_seconds_batch": 17.743225551908836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3148258924484253, "target_z": 0.41406246419191756, "ce": 6.1020402908325195, "genuineness": 4.2751336097717285, "identity": 0.14903989434242249, "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_TENS_reg_low/5786c25f105a0405ffbf86deb04795e15ea1ffbcf1000f91bd12a8fdb2b166a7.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": "57894a95f4ad46d84afb5e710645d6c5c49dcb6a790e7ba5638db2ff477c316c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a749d0d026ed7a5a740f4cb207bcf5bdc5f4bcbb0272d3a9f2a034513295c9f", "generation_seconds_batch": 21.8596139929723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7464081048965454, "target_z": -0.10449213826388666, "ce": 6.149696350097656, "genuineness": 4.921566009521484, "identity": 0.8188518285751343, "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_TENS_reg_low/57894a95f4ad46d84afb5e710645d6c5c49dcb6a790e7ba5638db2ff477c316c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "580046dfeadb2ca379686b25a9e51ca17faaa7ef90d6aab7a421daf64cf49c93", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5befdd795bbc3407005fd913959b0fe3cd48d139e2425583a8465409b8089adf", "generation_seconds_batch": 18.440039876033552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1536073684692383, "target_z": -0.5937501432323297, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.6887661218643188, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/580046dfeadb2ca379686b25a9e51ca17faaa7ef90d6aab7a421daf64cf49c93.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": "58751736afc6720e144c513c59f3d10bc5596fdaaaf2119c5c1f86db62f8f3cd", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "778817df21564ae804ff1f1e3e2b737759c317f6b91126a14637da4d3d63ab99", "generation_seconds_batch": 21.190567411947995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4716910123825073, "target_z": 0.22558590616792787, "ce": 5.959072113037109, "genuineness": 4.68307638168335, "identity": 0.7054238319396973, "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_TENS_reg_low/58751736afc6720e144c513c59f3d10bc5596fdaaaf2119c5c1f86db62f8f3cd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "5898dc10f3041cc833d7034cd2d650abd249d483c1547cb2accb2e272ec76ccb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "034775b08fd69ac6ddfec4724ab12d746d6634053d4944ce6dfb9fd4fde7db4c", "generation_seconds_batch": 25.997556727030315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.561095952987671, "target_z": 0.11816409383207213, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.6755775213241577, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5898dc10f3041cc833d7034cd2d650abd249d483c1547cb2accb2e272ec76ccb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "58a5be0e9c9221eeacaf44cdb298a32c43e364f36a2260f8953801e6be62c1fd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09fa61e953e4916af0ab2cc980c6a8cd7d4093adb6c51f25d9e3c15ba4e92e19", "generation_seconds_batch": 25.31049952202011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2446377277374268, "target_z": -0.7031249283838351, "ce": 6.221180438995361, "genuineness": 4.632867813110352, "identity": 0.4614325761795044, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/58a5be0e9c9221eeacaf44cdb298a32c43e364f36a2260f8953801e6be62c1fd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "58bf3559d617d4eb7a8c08ee71dafe547d09d8ffe8e8542650b587083bd7e4cc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7543282cc093f094be405f4e9e03a59b83d71e2fb5e5b975e7bc429fbdd865c", "generation_seconds_batch": 23.48091812990606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5326489210128784, "target_z": 0.15234376790404122, "ce": 6.2767791748046875, "genuineness": 4.18099308013916, "identity": 0.19058948755264282, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/58bf3559d617d4eb7a8c08ee71dafe547d09d8ffe8e8542650b587083bd7e4cc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "58e592fa6dfbc72829417144a2d80e2bd35ff044f46d1eb7520160a229674635", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "de09590649daa0e7034e711d42ef1243b68027074f6510d34a3653554e6b88c1", "generation_seconds_batch": 24.16671350202523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4397032260894775, "target_z": -0.9375, "ce": 6.125868320465088, "genuineness": 4.231201171875, "identity": 0.31354185938835144, "hypothesis": "Ik 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_TENS_reg_low/58e592fa6dfbc72829417144a2d80e2bd35ff044f46d1eb7520160a229674635.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "597696048ddad9788064ed69f1b06e47a8e2afa2bb65911c118990a7b0823b1c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fca1020d9e546ed426b44709e611d76a283128d5674a59b5e3779ce536a3233c", "generation_seconds_batch": 17.85253371600993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.017061471939087, "target_z": -0.42968753580808244, "ce": 6.133810997009277, "genuineness": 4.570107460021973, "identity": 0.5729842185974121, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaagt.", "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_TENS_reg_low/597696048ddad9788064ed69f1b06e47a8e2afa2bb65911c118990a7b0823b1c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "59a7808083d46359906059e62aff453c9b14b5d035f40a357e3c3e539f78047b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49f75781aeb1ca82cb7092455dbb08fb3da8c37447ea95de8ad7e122c9b61250", "generation_seconds_batch": 25.664999924018048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8853923082351685, "target_z": -0.271484419760103, "ce": 6.372091293334961, "genuineness": 4.293961524963379, "identity": 0.6277400255203247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/59a7808083d46359906059e62aff453c9b14b5d035f40a357e3c3e539f78047b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "59f88954a334652d721c05af73e5a934f2da5c4a39a527cb98b4c5f9030d7b2c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "af2baf26de26cda4aae2d1dcc8f4956bd4112b43fa8b6c81b7927f8b27f4dd40", "generation_seconds_batch": 26.47588178003207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.587917447090149, "target_z": 0.08593753580808243, "ce": 5.5817952156066895, "genuineness": 3.9425034523010254, "identity": 0.6353482604026794, "hypothesis": "La puerta sigue abierta. He dejado tu abrijo sobre la silla. Dime cuándo estás.", "wer": 0.13333333333333333, "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_TENS_reg_low/59f88954a334652d721c05af73e5a934f2da5c4a39a527cb98b4c5f9030d7b2c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "5a813a2416182a92e663dfd7b0d982553cc24593e6fd5262c8b83f809a264bd4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9f019e4f51eb8095c2cb78fb8767820a7b824063d487361986ff74a35b89189", "generation_seconds_batch": 26.748798663029447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061240792274475, "target_z": 0.71875, "ce": 6.117925643920898, "genuineness": 4.595211505889893, "identity": 0.5954246520996094, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5a813a2416182a92e663dfd7b0d982553cc24593e6fd5262c8b83f809a264bd4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "5af0cf8a536b50b7361e478c07d052ca5e50e183baff850aa216d795873dde40", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5a3ee3f621e3c92f7cae0ab2362c88d580a069e6cce75b3e8a7ee42260fc66b", "generation_seconds_batch": 24.08899778907653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361966609954834, "target_z": 0.35742195556818546, "ce": 5.871703147888184, "genuineness": 4.607763767242432, "identity": 0.7167181968688965, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5af0cf8a536b50b7361e478c07d052ca5e50e183baff850aa216d795873dde40.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "5afaf2cd7fc5f4ece03002a13651bb313802b0284d26e56a21f03ba5464c0bc8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7aac379fcffc8ec43177a88866a61b96e37388e1c470203b6aebfc12d4d9893e", "generation_seconds_batch": 25.31049952202011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7399059534072876, "target_z": -0.09667967407196909, "ce": 6.141753673553467, "genuineness": 4.494794845581055, "identity": 0.6077420711517334, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5afaf2cd7fc5f4ece03002a13651bb313802b0284d26e56a21f03ba5464c0bc8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "5b109f28fef7d9b676343f2b971942b2ea1b0a199685a4add3468618f3680393", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a9be95f60ac99d79c6f1dd0941ed43070a16990ac59cd0b2d379387c9d5a9e8", "generation_seconds_batch": 24.576194360037334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9910527467727661, "target_z": -0.39843753580808244, "ce": 5.935244083404541, "genuineness": 4.319066047668457, "identity": 0.5587837100028992, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5b109f28fef7d9b676343f2b971942b2ea1b0a199685a4add3468618f3680393.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "5b91e7f5dcc308a4f4b74ce7d07cec596d2ac0eb6f3baa1f6658e3f77bf82cf6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e1fa1b9cec396eecebc267eaa58ca623536df0817e67af75554f4242036f332", "generation_seconds_batch": 25.2665543720359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5562193393707275, "target_z": 0.1240234419760103, "ce": 6.308549880981445, "genuineness": 4.494794845581055, "identity": 0.7066901922225952, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5b91e7f5dcc308a4f4b74ce7d07cec596d2ac0eb6f3baa1f6658e3f77bf82cf6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "5c621c450f2bcd863843ad6663b07c64f052bfe79c4ad931f7b94c5ded6b4584", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "320852e8fac392b1b9c857f5588538a28e979170e248ba8db3fa2547f218242c", "generation_seconds_batch": 26.72806417301763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8122427463531494, "target_z": -0.18359376790404122, "ce": 5.895530700683594, "genuineness": 4.406930446624756, "identity": 0.2466796636581421, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5c621c450f2bcd863843ad6663b07c64f052bfe79c4ad931f7b94c5ded6b4584.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "5c9be575d160456a28839b5ecf2da5d92c85b2ae2e584da79add9393893db8ca", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "429900651ae0ce797a11ae42fb537dd8c3cfc2b42a78d100b61c916107057a6a", "generation_seconds_batch": 17.90274157503154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6078304052352905, "target_z": 0.06201172098800515, "ce": 6.324435234069824, "genuineness": 4.080576419830322, "identity": 0.3099115490913391, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5c9be575d160456a28839b5ecf2da5d92c85b2ae2e584da79add9393893db8ca.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "5cda778fbc1aacb63fb3d4179dc038d78dc2357adc81fc7c78e5c165f3a7dd78", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "b197d3aa4f701a0b2225dd3f648ee8e1c498b7f56c1aeb26d8ff231c54e6eee5", "generation_seconds_batch": 17.734317956026644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.780544638633728, "target_z": -0.14550786173611333, "ce": 5.959072113037109, "genuineness": 4.193545341491699, "identity": 0.5450075268745422, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5cda778fbc1aacb63fb3d4179dc038d78dc2357adc81fc7c78e5c165f3a7dd78.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "5cfebf5a826f390924be65d6a7f476f379644c807c6cc27c68cd70572a8fa37b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "61cb2f73a552161b745345018fcf8dd396fc24d175eee61c93539cdae0dc20c4", "generation_seconds_batch": 18.770202643005177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5318361520767212, "target_z": 0.1533203259280309, "ce": 5.998785495758057, "genuineness": 4.319066047668457, "identity": 0.7362041473388672, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5cfebf5a826f390924be65d6a7f476f379644c807c6cc27c68cd70572a8fa37b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "5d40a0060aa3e74b7e04ad965ca9a41942a9173794f08e6343d614612f558f8b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "20063618cb6fd26ff00dad7934f633f4b50c633163bacaf2513f44ad44368fb8", "generation_seconds_batch": 15.580047006951645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7707912921905518, "target_z": -0.13378902221590727, "ce": 5.5817952156066895, "genuineness": 3.8420867919921875, "identity": 0.6052590608596802, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5d40a0060aa3e74b7e04ad965ca9a41942a9173794f08e6343d614612f558f8b.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": "5d811b42c33ec1348827606c108a34514c7ca0c3d28dfd530cd19a13576444ae", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28657bcebde53a2cfa73abe8a6fc2424c344a38d2aa732aa4092c4b83986e29f", "generation_seconds_batch": 19.724639949039556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5834472179412842, "target_z": 0.09130860494002575, "ce": 6.221180438995361, "genuineness": 4.3881025314331055, "identity": 0.7628161907196045, "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_TENS_reg_low/5d811b42c33ec1348827606c108a34514c7ca0c3d28dfd530cd19a13576444ae.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very relaxed and unstrained. 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": "26c5af0f6cf8e0f106bf1b3bdc20744bf1a9a68ec138f268aa2dafa72a651b86", "generation_seconds_batch": 18.15688769193366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.787859559059143, "target_z": -0.1542968839520206, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.8513460159301758, "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_TENS_reg_low/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "5e68c76bbb7b540b7fef776d65a923c514e97adf64639dc89f45dc42e6f45b7d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e7701102fc01a25f183dfbe38d7e879a7333236eef8bebe0e6032ce6d05ce596", "generation_seconds_batch": 17.31516316498164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.426698923110962, "target_z": -0.9218750716161649, "ce": 5.895530700683594, "genuineness": 4.080576419830322, "identity": 0.6024516820907593, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5e68c76bbb7b540b7fef776d65a923c514e97adf64639dc89f45dc42e6f45b7d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "5eabe015e97bdbd70943d3699a6b1e1e5b74cda87efc1f8e69ada5554b923256", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58d16a640cf6f49ac88a2a611ea1576ee20fa5905a5ab105819b0452d5dc1b8d", "generation_seconds_batch": 22.812596139032394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6720393896102905, "target_z": -0.015136649371840295, "ce": 6.149696350097656, "genuineness": 4.080576419830322, "identity": 0.384784460067749, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5eabe015e97bdbd70943d3699a6b1e1e5b74cda87efc1f8e69ada5554b923256.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "5ed6f5148aaae6a06fc976108606f84dcd2c0f3149b9308abd6e6545cb29b1dc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "031d499b0f9b72a089eaf25fa31c48f447290eabf7b8103e5390d991b3c1b476", "generation_seconds_batch": 23.72884730703663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2901530265808105, "target_z": -0.7578124641919176, "ce": 6.109982967376709, "genuineness": 4.293961524963379, "identity": 0.5659588575363159, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5ed6f5148aaae6a06fc976108606f84dcd2c0f3149b9308abd6e6545cb29b1dc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "5ed790dc46093c26b6301f956439a6bbca2e5bca15bd401545427d64f7ba3aad", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86791aa71d8bb50b0f8cac15a54947cb1c2c65084b6f3a4e42e9c690dce45d71", "generation_seconds_batch": 21.130819741054438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5033891201019287, "target_z": 0.1875, "ce": 6.0305562019348145, "genuineness": 4.143336772918701, "identity": 0.6023520231246948, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/5ed790dc46093c26b6301f956439a6bbca2e5bca15bd401545427d64f7ba3aad.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "5fe0612eccd734b8e4936be593fcef8b85e01f19e0f45c1ac7d022da9925a44b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "81eec0a3f6d3c339d4a9daee830115a072a15d65c83221b43aef4f1d02957d2e", "generation_seconds_batch": 24.199689535889775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4221118688583374, "target_z": 0.2851562320959588, "ce": 6.300607204437256, "genuineness": 4.7081804275512695, "identity": 0.739712655544281, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/5fe0612eccd734b8e4936be593fcef8b85e01f19e0f45c1ac7d022da9925a44b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "604b9953f83a2dbe9efdcccd6ca154eece73d588dd51cbb52f2dd2fafe1e5a8a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27371c3794c564b249c61e7ed778047e98c912002d897b7bb293e67d3abdac27", "generation_seconds_batch": 23.92976638104301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6347687244415283, "target_z": -1.1718750716161648, "ce": 6.054384231567383, "genuineness": 4.118232727050781, "identity": 0.6910481452941895, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/604b9953f83a2dbe9efdcccd6ca154eece73d588dd51cbb52f2dd2fafe1e5a8a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "60b14554a30e54bfa9c56b8b5c6042cdf63d1363e8a313b6cfba21a4ffab0d09", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eb7d1209d41c503fa4572e3785bc92fca884c64519ff6325a22fe9249152cee", "generation_seconds_batch": 15.91721858305391, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417235255241394, "target_z": 0.291015580239897, "ce": 6.0305562019348145, "genuineness": 4.381826400756836, "identity": 0.598237931728363, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/60b14554a30e54bfa9c56b8b5c6042cdf63d1363e8a313b6cfba21a4ffab0d09.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "61cf86156e2ba9a747bf4bfc602706ef934ada4e2fcf5297a44076f2db389e6a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "6055d974f24f031163e5831bc86f055795573d44aff69ccb5c4b2e39fdc42622", "generation_seconds_batch": 19.09389638202265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3226640224456787, "target_z": -0.7968750716161649, "ce": 5.95112943649292, "genuineness": 4.193545341491699, "identity": 0.428470253944397, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/61cf86156e2ba9a747bf4bfc602706ef934ada4e2fcf5297a44076f2db389e6a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "61d84b297dd2fac5052b633dadb6c4996e5678de5f76f486e42b44fb21848025", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c563df5a6dd292a4c6e63d07d0ec7f982dc8a34e08a6119a1e75b0ac75c5421d", "generation_seconds_batch": 18.32566937804222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3505878448486328, "target_z": 0.3710937679040412, "ce": 6.348263263702393, "genuineness": 4.243753433227539, "identity": 0.5248998403549194, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/61d84b297dd2fac5052b633dadb6c4996e5678de5f76f486e42b44fb21848025.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "61e8d2a2411b1efa1857caf04084e372e1d9e6a18136d38fbdca3aad222adc6d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb161d04c217e41cc77e930bcd2f3dcc400f1a55108bc0d7fe601f532af7892f", "generation_seconds_batch": 26.291318524978124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6318072080612183, "target_z": 0.033203116047979396, "ce": 6.1894097328186035, "genuineness": 3.886019229888916, "identity": 0.46246156096458435, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/61e8d2a2411b1efa1857caf04084e372e1d9e6a18136d38fbdca3aad222adc6d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "627b5a7a0148ea8f2fc79d2fa65fd6d01ed1e626eba3d834d405c6e561c4ab32", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49c2293b104367f4faba2b6bd64182db721235bd9d4ecff498082f7905d352e3", "generation_seconds_batch": 18.364171614055522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4765676259994507, "target_z": 0.2197265580239897, "ce": 6.117925643920898, "genuineness": 4.331617832183838, "identity": 0.7495225667953491, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/627b5a7a0148ea8f2fc79d2fa65fd6d01ed1e626eba3d834d405c6e561c4ab32.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "635f17ad9004a8424f9ae0a00c2c3ffcbb66b0a54026de1f0f157b20f1af07c4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34322e0f581a0de0341ac331cad7608d473c00ee7e9dfe20281c272e712343f5", "generation_seconds_batch": 21.510994166019373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9000221490859985, "target_z": -0.28906246419191756, "ce": 6.165581703186035, "genuineness": 3.873466968536377, "identity": 0.32837358117103577, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/635f17ad9004a8424f9ae0a00c2c3ffcbb66b0a54026de1f0f157b20f1af07c4.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": "637912d917020a3c399db8b870a9aca6bdc507883f0466675b73f9fdb1f0cf8a", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56ec4deefbc07347dc44ed31f4ca650af8d3f4bfec2f94e188518638047fd345", "generation_seconds_batch": 20.51433985296171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4960741996765137, "target_z": 0.19628902221590727, "ce": 6.2767791748046875, "genuineness": 4.20609712600708, "identity": 0.46411532163619995, "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_TENS_reg_low/637912d917020a3c399db8b870a9aca6bdc507883f0466675b73f9fdb1f0cf8a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "64248c28ac3c2e751317ab73236711f5d7ab9fcc0abb676deab599e8896e46fb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "70ebe2cea77247a9766be56d2687ce336646be5dd7f9070b64dfcd4d22158622", "generation_seconds_batch": 15.580047006951645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5570321083068848, "target_z": 0.12304688395202061, "ce": 5.752562999725342, "genuineness": 4.362998008728027, "identity": 0.5824917554855347, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/64248c28ac3c2e751317ab73236711f5d7ab9fcc0abb676deab599e8896e46fb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "646cd39517637b623bd5a1daeeb90bdbcc2d43ec35027f4aaf5eeb03a605efe8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2dbefecc732c6ac3963e00f17964f3d06717477287fe184f291e3679438dfdc", "generation_seconds_batch": 24.842032745014876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4521844387054443, "target_z": 0.2490234419760103, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.7399985790252686, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/646cd39517637b623bd5a1daeeb90bdbcc2d43ec35027f4aaf5eeb03a605efe8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "647e05cb0c3d72ac983361822daccb82db26becbf537f8f00bc1facbc59b527d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e440f5c395fe6b5e26e65c50b56f5c75d2bb4935a33d85c155bafa71fa30711", "generation_seconds_batch": 26.362437876989134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3993542194366455, "target_z": 0.3125, "ce": 6.0305562019348145, "genuineness": 4.231201171875, "identity": 0.3909096419811249, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/647e05cb0c3d72ac983361822daccb82db26becbf537f8f00bc1facbc59b527d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "6516815b297773e38057d491f51a6f16b733cc5a9658a6f0f92ae474af897c57", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd8d9b2689b92b86135c0fdcf5bb59d5071c5127db5e8759e30655ea72af5d77", "generation_seconds_batch": 23.499189599999227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.630994439125061, "target_z": 0.03417967407196909, "ce": 6.078212261199951, "genuineness": 4.256305694580078, "identity": 0.5050419569015503, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6516815b297773e38057d491f51a6f16b733cc5a9658a6f0f92ae474af897c57.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "654a1469694d229cb801e83a9d8caf4f04430fc8ccc44b78ef199e0009dba7ae", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc15467ae189982b1380013e80a831f521ac5dff1e69925df0f8cfa19f2dd3ba", "generation_seconds_batch": 23.08492805901915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5017635822296143, "target_z": 0.1894531160479794, "ce": 6.1894097328186035, "genuineness": 4.457139015197754, "identity": 0.6398282051086426, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/654a1469694d229cb801e83a9d8caf4f04430fc8ccc44b78ef199e0009dba7ae.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "6578ae5310213196734348f78df3f3fde0fcc420efccfa4174c0bcb52012fdf8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ff54d11cb4afff15fe7d1e0546bab2f59d690e8341a9055d4a9afc89f6c35e30", "generation_seconds_batch": 16.350950531079434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6582223176956177, "target_z": 0.0014648370359845446, "ce": 6.197352409362793, "genuineness": 4.632867813110352, "identity": 0.7829070687294006, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6578ae5310213196734348f78df3f3fde0fcc420efccfa4174c0bcb52012fdf8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6588918a9ee0bdf296af97aeb846b346bf59c4de8e3fd9af857181ec2d855e93", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "f7edc124c607025c37e881b4279d0b160d3b300769267b6ec354e29e475663fd", "generation_seconds_batch": 17.211615551961586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8317493200302124, "target_z": -0.20703130371212364, "ce": 6.308549880981445, "genuineness": 4.281409740447998, "identity": 0.5890252590179443, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6588918a9ee0bdf296af97aeb846b346bf59c4de8e3fd9af857181ec2d855e93.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "65a6d87e10143b16e0d205e98e0169f2c30bb07f9add32a16e8948d96b79379c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83c1a45f725f4e13f43ec891bff44385e43d63f402ec9070976daee9f811f61e", "generation_seconds_batch": 16.074291587923653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4700654745101929, "target_z": 0.22753902221590727, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.788211464881897, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/65a6d87e10143b16e0d205e98e0169f2c30bb07f9add32a16e8948d96b79379c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "65b20b7dacef878bb4265a557ea34dbe426387beee1f6f126a25990c17ead947", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26af2352e8b13162a6f95c5eb00e1b1c4a5a309968a273591facc2f36a03b91a", "generation_seconds_batch": 24.576194360037334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.906524419784546, "target_z": -0.29687507161616483, "ce": 6.006728172302246, "genuineness": 4.65797233581543, "identity": 0.6304996609687805, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/65b20b7dacef878bb4265a557ea34dbe426387beee1f6f126a25990c17ead947.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "65be205690bb3a45d45113d4b999fd4a5d010df8eed5616a391d39d305ec5efd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc11e24457cd56f1b3262c25369c2b7a5e3a6bfe4be29f2843c5e3468b50952a", "generation_seconds_batch": 22.331478249048814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746814727783203, "target_z": -0.8593750716161649, "ce": 6.149696350097656, "genuineness": 4.419482707977295, "identity": 0.6746931672096252, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/65be205690bb3a45d45113d4b999fd4a5d010df8eed5616a391d39d305ec5efd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "66b245a8d2181e9f8d7f1dc93adc9f2bb1ed6c093440c8fde73a6aac33e3b73d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bde85a95b3cb34bb85670acb538b37d7807a8db3864fe91056c01d4b06ec44ee", "generation_seconds_batch": 23.595418605022132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070704460144043, "target_z": -0.4941406518560618, "ce": 6.260893821716309, "genuineness": 4.093128204345703, "identity": 0.2973877191543579, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/66b245a8d2181e9f8d7f1dc93adc9f2bb1ed6c093440c8fde73a6aac33e3b73d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "66dcbef865f04c471a25483a21502c6884929e9dfcc557766285f195a6c59853", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "433d954fd7f71072a33165cb3d9eb07cd37c34c5e52405e367ae356c366da290", "generation_seconds_batch": 24.14342634694185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6691946983337402, "target_z": -0.011718696287876357, "ce": 6.173524379730225, "genuineness": 4.68307638168335, "identity": 0.7875006198883057, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/66dcbef865f04c471a25483a21502c6884929e9dfcc557766285f195a6c59853.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "6729eba2c9adf7006ce322c1c91fd3b019c9cc3e1231d800bce6430fb9c893ff", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "f8e38efb63df01b33d346d8eddfe4e9c225e0b69c5d4d96fbf250d50c60c6be1", "generation_seconds_batch": 16.277138358913362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6917492151260376, "target_z": -0.038818396302085, "ce": 6.41180419921875, "genuineness": 4.193545341491699, "identity": 0.6134301424026489, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6729eba2c9adf7006ce322c1c91fd3b019c9cc3e1231d800bce6430fb9c893ff.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "6753ceeda3e9c1e44a50a92289fcda44e65824faed66996fb24e7b2ab04d123b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a51fff843eace4eb0c1486c9e1f599703838215d003c15bcaca92d9e2165786", "generation_seconds_batch": 21.130819741054438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.700080156326294, "target_z": -0.04882818766414425, "ce": 5.919358730316162, "genuineness": 4.532451152801514, "identity": 0.5435758829116821, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6753ceeda3e9c1e44a50a92289fcda44e65824faed66996fb24e7b2ab04d123b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "67bfa77407c376387a320df9d4e636c331c1d616fd4e8725899c700d9b88323b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9124a8a800769d041f04841f0ee5235be8ff9ba048f4077e58bc6126914a65a5", "generation_seconds_batch": 25.566488189040683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.166611671447754, "target_z": -0.6093750716161649, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.6909724473953247, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/67bfa77407c376387a320df9d4e636c331c1d616fd4e8725899c700d9b88323b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "67d4b3e2f1e3ea816f227348c608a6d98a2d4e8efb15ed757a7a2adba6554fc5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac1c9c73715758818009b93e9eb980cb318f212ae8b7ca79cd27f7b3c48e9255", "generation_seconds_batch": 22.87763513904065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6204283237457275, "target_z": 0.04687507161616486, "ce": 6.443574905395508, "genuineness": 4.7458367347717285, "identity": 0.40901049971580505, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/67d4b3e2f1e3ea816f227348c608a6d98a2d4e8efb15ed757a7a2adba6554fc5.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": "68b6d876e4d6f61541be78b6d6a00069bca4dc441b672ebea9632b756e5fb24d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "62e5515d0ae292d471b606166eb3881c25c0edf4e8603d3de1f2a153d62fa79c", "generation_seconds_batch": 28.483411198016256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9130265712738037, "target_z": -0.30468753580808244, "ce": 5.855817794799805, "genuineness": 4.6203155517578125, "identity": 0.6469907760620117, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/68b6d876e4d6f61541be78b6d6a00069bca4dc441b672ebea9632b756e5fb24d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "6988d26ba476718930912a155d5d41f3c9eebe66f7993398312e3ec69035cc01", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68855928c4985271835fab45df970dd9413730fd29d70ba9824c16bb2a5b3fbc", "generation_seconds_batch": 24.669813812011853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1327648162841797, "target_z": 0.6328124641919176, "ce": 6.022613525390625, "genuineness": 4.557555198669434, "identity": 0.727652370929718, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6988d26ba476718930912a155d5d41f3c9eebe66f7993398312e3ec69035cc01.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "69c4b26e773af9e2e0282ee750ebfc387d58b1acdb38ec5eee9a8b91e1c1d6f9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "141bb9fea2d48993a9595ec0302db1cf366a813ed066a9829d93864a4a798a6d", "generation_seconds_batch": 22.27617777499836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498222827911377, "target_z": -1.0078124641919175, "ce": 6.125868320465088, "genuineness": 4.400654315948486, "identity": 0.8127474188804626, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/69c4b26e773af9e2e0282ee750ebfc387d58b1acdb38ec5eee9a8b91e1c1d6f9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "69f110f53ead29e06068eaea298c9e1c795fdcb8085390a3286d023802c64562", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "90dfa97fde91247f1f60d7d3cabfed2847cb59b31a19efe10e5bf93c394546c3", "generation_seconds_batch": 19.41557416692376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2999062538146973, "target_z": -0.769531160479794, "ce": 5.752562999725342, "genuineness": 4.080576419830322, "identity": 0.5804261565208435, "hypothesis": "De deur 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_TENS_reg_low/69f110f53ead29e06068eaea298c9e1c795fdcb8085390a3286d023802c64562.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very relaxed and unstrained. 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": "088fc3e7cc859902fb68ce218e007e028bcda3e7e9fe0611335abf6da202d242", "generation_seconds_batch": 17.288285846007057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8415025472640991, "target_z": -0.21875, "ce": 6.308549880981445, "genuineness": 3.9048471450805664, "identity": 0.23038609325885773, "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_TENS_reg_low/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.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": "6a91c2e24c4afbe31edd5f02055e2c08ee44b6ffdce05418d67a08b94aa753be", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bb6597603cb986085775348e624fd0b84c71434517bc3e68c8e4df809302026", "generation_seconds_batch": 24.16592906008009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": -0.644531160479794, "ce": 6.165581703186035, "genuineness": 4.607763767242432, "identity": 0.5617632865905762, "hypothesis": "Je berucht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6a91c2e24c4afbe31edd5f02055e2c08ee44b6ffdce05418d67a08b94aa753be.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "6aa4a4ccc155d1c5de08ee2a1607dd313a2a46dd34dc22bd6cb0c53b745b1b27", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "35233d756c66ad0dfefca6ba33433fe109776c871b5966d485bfdf0d33ae49f4", "generation_seconds_batch": 23.92186559503898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6374965906143188, "target_z": 0.026367209880051517, "ce": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.6008855700492859, "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_TENS_reg_low/6aa4a4ccc155d1c5de08ee2a1607dd313a2a46dd34dc22bd6cb0c53b745b1b27.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "6abc4868c81831671b333f8ef9389b557e91988694f3b87737b660dc9ff1824d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95fa5ddf2faa3fe3aa64a19356f64f6c204aaf2d05a952ea5cc6e811d136b399", "generation_seconds_batch": 25.18278258200735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.40097975730896, "target_z": 0.3105468839520206, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.38365840911865234, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6abc4868c81831671b333f8ef9389b557e91988694f3b87737b660dc9ff1824d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "6ad6ccf67975cfc258aa1702b6024933c8fcfa9bba084eb9c1f5df4f4e3a3329", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c9c5e4f4aad35d55f1b08093f6b930e3102e1c7b3bfb54350831ee8bb034d89c", "generation_seconds_batch": 17.962101444019936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3603410720825195, "target_z": 0.35937507161616483, "ce": 6.229123115539551, "genuineness": 4.406930446624756, "identity": 0.6412872076034546, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6ad6ccf67975cfc258aa1702b6024933c8fcfa9bba084eb9c1f5df4f4e3a3329.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6ae9fb56e5ccdb111f0f169da67d5f72f1527893b66585754c411a58eb587138", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9fa62f131e1aedadf42e3d3014de7414f7db9cab1704039e48752dc3e9571a7e", "generation_seconds_batch": 23.838397464016452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8366259336471558, "target_z": -0.21289065185606182, "ce": 5.935244083404541, "genuineness": 4.250029563903809, "identity": 0.6531317234039307, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kaart.", "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_TENS_reg_low/6ae9fb56e5ccdb111f0f169da67d5f72f1527893b66585754c411a58eb587138.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very relaxed and unstrained. 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": "e0f417291572c45a046230aa7767f62af03855c805d463443c774068866cd3be", "generation_seconds_batch": 20.559019336942583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6988608837127686, "target_z": -0.04736320739582999, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.7038094997406006, "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_TENS_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": "6bb0ee8343099010e786711308d2cb38268320f9b8a5c80340f112d45a9f5448", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "01d4003266aac87690b73339facf150fdf9af5edc6357cbfbc37faafc0e4a03f", "generation_seconds_batch": 18.63830425508786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6131134033203125, "target_z": 0.055664093832072124, "ce": 6.0305562019348145, "genuineness": 4.557555198669434, "identity": 0.6382125616073608, "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_TENS_reg_low/6bb0ee8343099010e786711308d2cb38268320f9b8a5c80340f112d45a9f5448.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "6bbe59871e114e028282d0536915f693e2e83a8b2c1a2fffccdc335d7b69c4d7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9fdae12e461213352dfd1c9c69258168e838692335a38e672d1285a61a77de5b", "generation_seconds_batch": 25.694240855053067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8545069694519043, "target_z": -0.23437507161616486, "ce": 5.919358730316162, "genuineness": 4.2751336097717285, "identity": 0.5084334015846252, "hypothesis": "I found the little box. It was beside the old window. We can look inside", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6bbe59871e114e028282d0536915f693e2e83a8b2c1a2fffccdc335d7b69c4d7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "6be7da7b68c14b3895cac81398339a6a13e82d4817618d3aa0b0e551e515227e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44b25c70850bcff749c11b73bec5dfcd6803fc32314f3110c5c22680bb53c90d", "generation_seconds_batch": 23.43631063203793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6687883138656616, "target_z": -0.01123041727588151, "ce": 6.213237762451172, "genuineness": 4.469690799713135, "identity": 0.7735841274261475, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6be7da7b68c14b3895cac81398339a6a13e82d4817618d3aa0b0e551e515227e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "6bf3daa877675acda8331e2f28b08f0e8953dec30bcacac6ac9188c230c66b9b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd0059cbe1d0dec50630e86a24888a58149b191f10780029e6f00fb1b1e656cf", "generation_seconds_batch": 25.66421750409063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6659436225891113, "target_z": -0.007812464191917571, "ce": 5.911416053771973, "genuineness": 4.080576419830322, "identity": 0.5941835641860962, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6bf3daa877675acda8331e2f28b08f0e8953dec30bcacac6ac9188c230c66b9b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "6c8a6b93c84ff77967747713b416fc432852d01ae79658d3079e355d068f729d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0d996805fa400c328dc4fde79b5e6fef22f4846766ff08ed11381514a0f8938", "generation_seconds_batch": 25.97847821598407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.251429557800293, "target_z": 0.490234419760103, "ce": 6.006728172302246, "genuineness": 4.331617832183838, "identity": 0.5737577080726624, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6c8a6b93c84ff77967747713b416fc432852d01ae79658d3079e355d068f729d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "6ca3475c880e612b554df2eb333d1ca382378444900c26f89710294d7a8fde2a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "465acb68300e7c62d0e9c92e63be2d8df0fc91c2e9e849c752125c4b33f29766", "generation_seconds_batch": 16.892138245049864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8398770093917847, "target_z": -0.2167968839520206, "ce": 6.292664527893066, "genuineness": 4.306513786315918, "identity": 0.3901718258857727, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6ca3475c880e612b554df2eb333d1ca382378444900c26f89710294d7a8fde2a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "6cc0a2c4e5ad393afc5bdda1946f98eb874f0493297773f7b8733c8a36ac5c1d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1d0a81467b46d5ba51e51655d4847ed477ba1c8ddf225a581b9e159040364aa", "generation_seconds_batch": 17.0072726609651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6196155548095703, "target_z": 0.047851629640154554, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.6188966631889343, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6cc0a2c4e5ad393afc5bdda1946f98eb874f0493297773f7b8733c8a36ac5c1d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6da438b12b7906eb678537153afa20dc21a6f36e6e143dc1bb9cf9c59762f729", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c00d9ccb5f0570f1cbfb41d433d8bd8a0c128b896a73fb756ab379815c950dc4", "generation_seconds_batch": 22.30804902408272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127598524093628, "target_z": -0.5625, "ce": 6.24500846862793, "genuineness": 4.18099308013916, "identity": 0.684828519821167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6da438b12b7906eb678537153afa20dc21a6f36e6e143dc1bb9cf9c59762f729.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": "6df031e2abc0c975a5942aa1d36b95d08547338d5fbc6cb5b92188713f98c61c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "43545698747c22eb9e2f88b261d7cb762844bba939a4c40ba4c097b1a0506679", "generation_seconds_batch": 22.970506611978635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8195576667785645, "target_z": -0.1923827901199485, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.6520686149597168, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel maar wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6df031e2abc0c975a5942aa1d36b95d08547338d5fbc6cb5b92188713f98c61c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6dfd328fbb2049b68c106661589dd89ec8ebddc6eed4985e789e7962777668db", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6296bcbe7aef76e36f03ed86e65ccb021ef107a7b6ec76cca5e4d428208ffce1", "generation_seconds_batch": 19.778386716963723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7049567699432373, "target_z": -0.05468753580808243, "ce": 6.086154937744141, "genuineness": 4.356721878051758, "identity": 0.3895873427391052, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "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_TENS_reg_low/6dfd328fbb2049b68c106661589dd89ec8ebddc6eed4985e789e7962777668db.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "6e7f8faac2667ad530d23099296872cb330c9023264d484fa1aeadf2c54b2f0e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f96ce9dc6585a16120a1737e5a6ca345ba9881f5d4427c6a554f8d2f19f654b", "generation_seconds_batch": 23.30500212602783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.034942388534546, "target_z": -0.45117181233585574, "ce": 5.4825119972229, "genuineness": 4.068024158477783, "identity": 0.671913743019104, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6e7f8faac2667ad530d23099296872cb330c9023264d484fa1aeadf2c54b2f0e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6f50d4f6f7c42ad03e080dacf845407a968f58cb1a9a8f05164b6b720f7ff948", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "747a12d0e16dedfdc76be15bf93cc0dfebcb061fe7acbb46157193fc57c942ae", "generation_seconds_batch": 24.897381274960935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7293399572372437, "target_z": -0.08398441976010304, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.8051318526268005, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6f50d4f6f7c42ad03e080dacf845407a968f58cb1a9a8f05164b6b720f7ff948.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "6f58e7c80f57ab6009cc61acc742bc094a84ce726628f221acc0249966663ab6", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "8699fec92ded2f9c18cb1da47b2942485eec8600f731a445d3c1e85803ddff3d", "generation_seconds_batch": 18.931981042027473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4749420881271362, "target_z": 0.2216796740719691, "ce": 6.348263263702393, "genuineness": 4.632867813110352, "identity": 0.7932111024856567, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/6f58e7c80f57ab6009cc61acc742bc094a84ce726628f221acc0249966663ab6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "6f86de8069c968812759cc8405785404faf5505f47abf7a25e384503b4c5e646", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d61d9f967748448075c2898c2f3a5e9ca0f5a8bdd4b0c8cb474c3f3269626ea7", "generation_seconds_batch": 18.61055895802565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7159291505813599, "target_z": -0.06787106913194334, "ce": 6.324435234069824, "genuineness": 4.281409740447998, "identity": 0.6683216094970703, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6f86de8069c968812759cc8405785404faf5505f47abf7a25e384503b4c5e646.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "6f98eb93c8e073b547e7ba90f76bf4d628cf9b79c401bbfe45b8754cf65b7e6e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "591926efa4c8bd257a7a85fe3d5363b52021d858b468ec65e9f7f5ef1e7f6b9d", "generation_seconds_batch": 16.708997096982785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4944486618041992, "target_z": 0.19824213826388667, "ce": 6.157639026641846, "genuineness": 4.570107460021973, "identity": 0.7574601173400879, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6f98eb93c8e073b547e7ba90f76bf4d628cf9b79c401bbfe45b8754cf65b7e6e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "6fcefa86d63450664f99039a5c713b9688a17ebdf986665e479018029be19409", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "928b3fe06ddfb36db5eb58ee1f5c722b19b3a66d07a07c116d4737b977707b8c", "generation_seconds_batch": 25.775045663001947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3635921478271484, "target_z": 0.3554688395202061, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.29677796363830566, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6fcefa86d63450664f99039a5c713b9688a17ebdf986665e479018029be19409.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "6fd2443b6e925035268ea6d4039311c1bbd8df89dc9a294f2ecc5cf41ad5cf44", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a191961cfac62ca488b0bdd75b4576474ed2d5581cdcb74dd41eb1826a6a9907", "generation_seconds_batch": 23.509477789979428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5139551162719727, "target_z": 0.17480474568813395, "ce": 6.2767791748046875, "genuineness": 4.394378662109375, "identity": 0.5030585527420044, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6fd2443b6e925035268ea6d4039311c1bbd8df89dc9a294f2ecc5cf41ad5cf44.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "6fd5cc8469a0840b553daa8b17ae28117d5b292f7e6b720f37ccd3a5b05de8a3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8865324fb7e06530b6bd3ebe4d1fc9ceee6ab77ab2adf815a54c6ef32cc8eab3", "generation_seconds_batch": 24.416452731005847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5814152956008911, "target_z": 0.09375, "ce": 6.117925643920898, "genuineness": 4.632867813110352, "identity": 0.6319464445114136, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6fd5cc8469a0840b553daa8b17ae28117d5b292f7e6b720f37ccd3a5b05de8a3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "6fe0ce8253aa02732153f8ca6166ac563a831cd3ae004a3a0ba99ccea038b77f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "518ac3d1a63517ca8a6cbfec925b5711696f4e32f9442fcf78e71bc335ac1fca", "generation_seconds_batch": 22.755767591064796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2121267318725586, "target_z": -0.6640623209595878, "ce": 6.1894097328186035, "genuineness": 4.344170093536377, "identity": 0.8413957357406616, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/6fe0ce8253aa02732153f8ca6166ac563a831cd3ae004a3a0ba99ccea038b77f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "70608669ff8b6c70681d118cfb14fda6f5b72753180a1ae24c43fe4fe8189c58", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c92c81e6a29d335bec048fefdfa89ae14dab4c162bdf3f02957b0bc643685765", "generation_seconds_batch": 17.112515517976135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7195866107940674, "target_z": -0.07226558023989696, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.34670037031173706, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/70608669ff8b6c70681d118cfb14fda6f5b72753180a1ae24c43fe4fe8189c58.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "7086751c939161fd806ce87c2ca4bca31a9701eea5225ee098e979a30c820aa4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "375caae8f19bc7a670680f3289f85be914fb24143e86d5534c73f01ab4747d1e", "generation_seconds_batch": 23.43836496397853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5334616899490356, "target_z": 0.1513672098800515, "ce": 6.133810997009277, "genuineness": 4.105680465698242, "identity": 0.5670285224914551, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7086751c939161fd806ce87c2ca4bca31a9701eea5225ee098e979a30c820aa4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "70aaaa759c969459ea74b6f6c328aa07acb53240cd0c2d86e9c4300533150d75", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "6a7810af618160f410f334069cf923cb7f7f279cd8cd3d1480dd78d4e7a1d713", "generation_seconds_batch": 17.016220916062593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.500950813293457, "target_z": 0.1904296740719691, "ce": 6.09409761428833, "genuineness": 4.406930446624756, "identity": 0.4885932505130768, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/70aaaa759c969459ea74b6f6c328aa07acb53240cd0c2d86e9c4300533150d75.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "711c3b0f339524ec518e297f0540054b3bf174e33cf7425b5885a0d7e8cf4aa5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b47fb851db17946c0939ba8dd964cfe542438543efc93a4875ee68ab2a713f94", "generation_seconds_batch": 19.338564528967254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6350582838058472, "target_z": 0.029296883952020607, "ce": 5.935244083404541, "genuineness": 4.218649387359619, "identity": 0.6826454997062683, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/711c3b0f339524ec518e297f0540054b3bf174e33cf7425b5885a0d7e8cf4aa5.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": "7169627d7f2aa2540414cce7061abd2a3572f4fb56d959112ac2823895f2e832", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d7f7e24248fec752cd1405443c57da6c29c2e51ad764dbd0e55a6f50d5c735b8", "generation_seconds_batch": 24.477823984925635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4481205940246582, "target_z": 0.2539062320959588, "ce": 6.0146708488464355, "genuineness": 4.758388519287109, "identity": 0.7815830707550049, "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": 0.120000000000001, "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_TENS_reg_low/7169627d7f2aa2540414cce7061abd2a3572f4fb56d959112ac2823895f2e832.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "719428278667b02d0f6aa6d24b8ec18cf0a3ece96159ec067a71d1be4b6d9928", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "309b978114484f0497d8038940e30ca0f4b3abc67afa9ca92ba8a9c1f6f6df59", "generation_seconds_batch": 23.84099049901124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4026052951812744, "target_z": 0.3085937679040412, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.542447030544281, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/719428278667b02d0f6aa6d24b8ec18cf0a3ece96159ec067a71d1be4b6d9928.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": "71d00b958c3ba104df8cc4359c727918d7475f600615168db8aa5c404300f6f0", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a31184190902f768a4301ef2371fc81bbd0799837b6c4c189f4de4620e409a8b", "generation_seconds_batch": 24.09276328701526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.577351450920105, "target_z": 0.09863279011994848, "ce": 6.173524379730225, "genuineness": 4.13078498840332, "identity": 0.640496015548706, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/71d00b958c3ba104df8cc4359c727918d7475f600615168db8aa5c404300f6f0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "71ff891e107c841e578d99427aa103aaf9eb92f8600718779d0981c6ce05cb59", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "564d4c23762e91c6719dc5d479c8f02fafe1e3ecb6ca8dce3b8e2e1748cad502", "generation_seconds_batch": 22.27617777499836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4852185249328613, "target_z": -0.9921875358080824, "ce": 6.181467056274414, "genuineness": 4.413206577301025, "identity": 0.5727143883705139, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/71ff891e107c841e578d99427aa103aaf9eb92f8600718779d0981c6ce05cb59.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": "7253cec26f4280399986e6a17803f3b3225a57c355bc2eee79f193d68af8d3f5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71f0b7f1e161e5c4beed36f8924bfa11e406fd04367d3889baa79d255c2b0408", "generation_seconds_batch": 19.858084617066197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7049567699432373, "target_z": -0.05468753580808243, "ce": 6.054384231567383, "genuineness": 4.7458367347717285, "identity": 0.8497756123542786, "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_TENS_reg_low/7253cec26f4280399986e6a17803f3b3225a57c355bc2eee79f193d68af8d3f5.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": "72953a4b922d0affcd456f547874970bc3cc3a4adfee001c5e6972cf47dd71aa", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d169e94fbfd68a880c54af0321a961baace04d9eb9fa184bb81e2819764abb2", "generation_seconds_batch": 20.527365706977434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7992383241653442, "target_z": -0.16796869628787636, "ce": 5.728734970092773, "genuineness": 4.3881025314331055, "identity": 0.6275321245193481, "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_TENS_reg_low/72953a4b922d0affcd456f547874970bc3cc3a4adfee001c5e6972cf47dd71aa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "72c8a85b33b2f100b6d0da0047977a54a788fe82327603c18421140afa63ad2d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b13a12ff9c1dcb2fbde2dab776d648e5624fddb18d275e153ae07e790361f300", "generation_seconds_batch": 24.01814999501221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0625767707824707, "target_z": -0.48437507161616483, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.7029851675033569, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/72c8a85b33b2f100b6d0da0047977a54a788fe82327603c18421140afa63ad2d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "72e474d94f08a7ac36664f5f652ddc7503e0cb0866973a963610538cc00f7c71", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c43ae07f528f7985d57064c3bc43efb75a47b04433edcc39c2164263247fcfc5", "generation_seconds_batch": 23.77860659500584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6618797779083252, "target_z": -0.0029296740719690893, "ce": 5.6413655281066895, "genuineness": 4.168440818786621, "identity": 0.44570428133010864, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/72e474d94f08a7ac36664f5f652ddc7503e0cb0866973a963610538cc00f7c71.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "730235ca002a8ba7e5b942772911007e631cfa751c3b1bd7f6610dcbc1eb0814", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2cf7681aff1b9aabb0b47f6b7c6b1171e605651c0680e7e674c6e191bfd8675", "generation_seconds_batch": 23.565933825913817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.856945276260376, "target_z": -0.23730474568813395, "ce": 6.086154937744141, "genuineness": 4.557555198669434, "identity": 0.7209143042564392, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/730235ca002a8ba7e5b942772911007e631cfa751c3b1bd7f6610dcbc1eb0814.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "733e14daa0e7a3ea099067cc000d0efcdd3cefb25eeb25a373bd3cf37877858a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7d6f58427eb759a445cc88e48ba7a513343cdfe5bd9be1dc8f0886dbf551e0ca", "generation_seconds_batch": 19.778386716963723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5480916500091553, "target_z": 0.13378902221590727, "ce": 6.268836498260498, "genuineness": 3.986435890197754, "identity": 0.6870709657669067, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in komen.", "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_TENS_reg_low/733e14daa0e7a3ea099067cc000d0efcdd3cefb25eeb25a373bd3cf37877858a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "735f3857b2aef5ba1b03b5ccfa2d13bcc2ac9567cf28b5abdaf598a1ad74efe9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0207b8cbe3eeacf92f3e7379781fbe9f769c171dc95997c92059057de07bae4d", "generation_seconds_batch": 23.949956815107726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5806025266647339, "target_z": 0.0947265580239897, "ce": 6.086154937744141, "genuineness": 4.13078498840332, "identity": 0.7169559001922607, "hypothesis": "Ik 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_TENS_reg_low/735f3857b2aef5ba1b03b5ccfa2d13bcc2ac9567cf28b5abdaf598a1ad74efe9.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": "73a3222e7e90de8c18e88cd4f8a0dca686e86ac81b16c04fd8a492c97e7802b1", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "6105f0c5f3ff925a8a31c7e7f639ac882782d9dbb73e2af98ef17a8d0565795a", "generation_seconds_batch": 17.678057114011608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9276564121246338, "target_z": -0.322265580239897, "ce": 6.459460258483887, "genuineness": 4.545003414154053, "identity": 0.198734849691391, "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_TENS_reg_low/73a3222e7e90de8c18e88cd4f8a0dca686e86ac81b16c04fd8a492c97e7802b1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "73a99be1872cd766c187636589ea7984ddb408db11297531bd24b6bcc724040b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b50057de29539a029378cb56020faa958046e342ff4983edf508ae7669e199e", "generation_seconds_batch": 24.14342634694185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5326489210128784, "target_z": 0.15234376790404122, "ce": 6.1020402908325195, "genuineness": 4.293961524963379, "identity": 0.5826349854469299, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/73a99be1872cd766c187636589ea7984ddb408db11297531bd24b6bcc724040b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "7417b9680166f23d1b37bacffd357d07dbea4d1775357cc9c435abbafafe0c7e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5c4da6015c94b80bffa999f6d4641fa57c978c888acd709dccc0335a2c4e2b6", "generation_seconds_batch": 24.49595363799017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3733454942703247, "target_z": 0.34375, "ce": 6.213237762451172, "genuineness": 4.7081804275512695, "identity": 0.6596145629882812, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7417b9680166f23d1b37bacffd357d07dbea4d1775357cc9c435abbafafe0c7e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "7423ca75987f89a51634c954b60f4c1176c9f0fd091f8b840a11234bba875543", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e42dd69369d5a78dbca3ced0d11daf1f17597bb36f0c626e25e2312edd5cdc9f", "generation_seconds_batch": 23.48091812990606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3944776058197021, "target_z": 0.3183593481439382, "ce": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.44407498836517334, "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_TENS_reg_low/7423ca75987f89a51634c954b60f4c1176c9f0fd091f8b840a11234bba875543.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "7477ed58061f8527a9af9b560b7de2ff5566f9659e3c19c30250038e645ecc34", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2b07605ab2b8c42b00c148984ed7860445effb8283b4c60b8662220f90ca981e", "generation_seconds_batch": 17.06571057497058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6602542400360107, "target_z": -0.0009765580239896964, "ce": 5.982900142669678, "genuineness": 3.9801597595214844, "identity": 0.5934665203094482, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7477ed58061f8527a9af9b560b7de2ff5566f9659e3c19c30250038e645ecc34.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "748daee4d7f2b50b786189ca33be67a3ba2d2401e89e012ea16074a3386fc356", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90c1b5352cdcf8500c373e8b1ae87f2c8ac8673e45de342bb94e209ddc11c172", "generation_seconds_batch": 18.12475952494424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6033601760864258, "target_z": 0.06738279011994848, "ce": 6.173524379730225, "genuineness": 4.093128204345703, "identity": 0.246900275349617, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/748daee4d7f2b50b786189ca33be67a3ba2d2401e89e012ea16074a3386fc356.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "74fb1749313b220eb9735a2a0eea73c65891112986b1980a7d69a91920785405", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "148cb25f671107b1cbfeb1fff44ec28db1d6f4b0a5bf3dd9c5125e39f1cdc112", "generation_seconds_batch": 16.590158174047247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.238135576248169, "target_z": -0.6953124641919176, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.590701699256897, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/74fb1749313b220eb9735a2a0eea73c65891112986b1980a7d69a91920785405.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "750da535e3fd451cdefb52be442b4775ac9a738a001689bf89da37a77262a472", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7ba0f61ebdc1c4563c80f11b7ceb0c28ee4c054646bec19124381fce27b70fbc", "generation_seconds_batch": 17.912434212048538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.852068543434143, "target_z": -0.23144525431186605, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.5795674920082092, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/750da535e3fd451cdefb52be442b4775ac9a738a001689bf89da37a77262a472.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "7517ea4129eed66aa7f64317294a564f7da4931585db40f6c2842b8f1bc2628c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56f312734b90d5c85e6c62a3ab375b44580093f8e75a55365fa2e74dde6b08cf", "generation_seconds_batch": 24.442128799972124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4091074466705322, "target_z": 0.30078130371212364, "ce": 6.038498878479004, "genuineness": 4.444586753845215, "identity": 0.6490879058837891, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7517ea4129eed66aa7f64317294a564f7da4931585db40f6c2842b8f1bc2628c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "753aa4cdee8a61e2b69729bac694a0fde315ecd45b67ebcbca7864c2344e2cc1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eec959df77868a548144e86a9e1797a11499bf5f06845b75c1f5a6222eb531c2", "generation_seconds_batch": 18.672299705911428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5627214908599854, "target_z": 0.11621097778409273, "ce": 6.197352409362793, "genuineness": 4.331617832183838, "identity": 0.4516940116882324, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/753aa4cdee8a61e2b69729bac694a0fde315ecd45b67ebcbca7864c2344e2cc1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "757e6bdd5cd39b8175c1e062cbacbb8c594025c8d0caac9e902306aa38038133", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba33877f7da6d1bfce4089c59e3a725d3c2397155cdd5abd9e350da125a6b5f6", "generation_seconds_batch": 26.09549324598629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3310813903808594, "target_z": 0.3945311604797939, "ce": 5.538110733032227, "genuineness": 4.632867813110352, "identity": 0.7209795713424683, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/757e6bdd5cd39b8175c1e062cbacbb8c594025c8d0caac9e902306aa38038133.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "758d3d42b1e3095fa690f4ab99c60b6087a2cb0d830c24a15ad58362c2fad3a9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8abc72d05c21f811683028afbc26b696f0b2e3c46065d84903afaa74fab4433", "generation_seconds_batch": 18.266433548997156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4830697774887085, "target_z": 0.21191409383207213, "ce": 5.95112943649292, "genuineness": 4.325342178344727, "identity": 0.681011438369751, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/758d3d42b1e3095fa690f4ab99c60b6087a2cb0d830c24a15ad58362c2fad3a9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "76152241cb1d465df0e41473c3acb571586a807d4dac77fc68a5280897821c8d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0bc7e0ec805736808d36086c61b731fcb36f785e86362b56b24b8a9bfe92d014", "generation_seconds_batch": 18.967712357989512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5777578353881836, "target_z": 0.09814451110795364, "ce": 6.038498878479004, "genuineness": 4.105680465698242, "identity": 0.6387553215026855, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/76152241cb1d465df0e41473c3acb571586a807d4dac77fc68a5280897821c8d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "76adf4fdbb4a61bc082d710c83d6ed772b6e1e9c9a216406546c208287f7647f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cce908c5afe4fa9d53c90b3b694a29e85ad2f001e544729b08007181375075f", "generation_seconds_batch": 24.10760329803452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595572471618652, "target_z": 0.4804688395202061, "ce": 6.181467056274414, "genuineness": 4.6203155517578125, "identity": 0.6245055198669434, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/76adf4fdbb4a61bc082d710c83d6ed772b6e1e9c9a216406546c208287f7647f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "76dea81efca4d0c89ec0de1196f49b41dcc4ba28118f8b7560cc4ecf0f731fa7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2511498cc1083b628b0d19ce327ec354e018377ffedff1a5ff66ff2ea2ea9b94", "generation_seconds_batch": 22.36469458299689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4879463911056519, "target_z": 0.20605474568813395, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.6553955078125, "hypothesis": "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_TENS_reg_low/76dea81efca4d0c89ec0de1196f49b41dcc4ba28118f8b7560cc4ecf0f731fa7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "772d9c2a95cb0e3c179fe9deaad0cf6a37fdbc8c86fb2c11d13c025189e165aa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b547b7c16fe2772c50671e05191bc65ea14b432398b71eb8499f0ecbdab05153", "generation_seconds_batch": 24.95824826299213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3733454942703247, "target_z": 0.34375, "ce": 5.863759994506836, "genuineness": 4.482243061065674, "identity": 0.8695059418678284, "hypothesis": "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_TENS_reg_low/772d9c2a95cb0e3c179fe9deaad0cf6a37fdbc8c86fb2c11d13c025189e165aa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "7743e914208536e2c7ef7f36791b911844972d26cf6f17c964d3c9942a3d17aa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9355b97a064bf896185bcf423e81e172f4b686ec2423cc81d0de9669bd20cf51", "generation_seconds_batch": 26.74374882504344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5448405742645264, "target_z": 0.13769525431186605, "ce": 5.967014789581299, "genuineness": 3.9425034523010254, "identity": 0.3158825635910034, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen inhouden.", "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_TENS_reg_low/7743e914208536e2c7ef7f36791b911844972d26cf6f17c964d3c9942a3d17aa.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very relaxed and unstrained. 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": "85ef5c87d5bc12915d6fb43f262cbdf26918741c7b5e7d46d32dd6c688d8aa41", "generation_seconds_batch": 18.647080836002715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7033312320709229, "target_z": -0.052734419760103034, "ce": 6.165581703186035, "genuineness": 4.595211505889893, "identity": 0.7713711261749268, "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_TENS_reg_low/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.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": "77c8bf9df35476409ed69f2de7ba15a777b747c32d9075f9df3e22d1411cdab2", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8e55b949483e4a893e18d7af5db6a114286851bde978ffc7353624f723fefd63", "generation_seconds_batch": 17.38362170290202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.618802785873413, "target_z": 0.04882818766414425, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.7481427192687988, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/77c8bf9df35476409ed69f2de7ba15a777b747c32d9075f9df3e22d1411cdab2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "77e4ca0c4fca7e95d96f7c174fcec9191f3a297f0c4e01983b4653facc5bfd27", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d99c104452be0331affe1e913ae4af17ec03d106f1e0b57e21a24772808a8b21", "generation_seconds_batch": 23.66779607301578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3343324661254883, "target_z": 0.39062492838383517, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.6500800251960754, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/77e4ca0c4fca7e95d96f7c174fcec9191f3a297f0c4e01983b4653facc5bfd27.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "77ffc78abe616cd3b75fc352ee72b0215aa60a7077b233bee8c5b2d3300a40e9", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "164e54728283d331d028d9026931897813613800bb9b51503c7b610750890baa", "generation_seconds_batch": 16.984286742983386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7854212522506714, "target_z": -0.1513672098800515, "ce": 6.348263263702393, "genuineness": 4.042920112609863, "identity": 0.5338141918182373, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/77ffc78abe616cd3b75fc352ee72b0215aa60a7077b233bee8c5b2d3300a40e9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "788b6b0a5726f9f145fe70d6ce21f511e7be2168448d9c334b042fc2422607c1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "4053ee2a8908442cb209db43632f67bb54a84ad89ec47dc002f99373543dc0d1", "generation_seconds_batch": 16.708997096982785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4985125064849854, "target_z": 0.19335934814393818, "ce": 6.173524379730225, "genuineness": 4.582659721374512, "identity": 0.816537618637085, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/788b6b0a5726f9f145fe70d6ce21f511e7be2168448d9c334b042fc2422607c1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "789efc6c7a9686676edac7c87b3af6a2e378d0812ee281fbbb65b550fba6e49f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24cd60f8f576b2ced84506517b1bbbbad8c3fb8db1fc41d2797d550e8db0f1e3", "generation_seconds_batch": 21.851931801997125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5521554946899414, "target_z": 0.12890623209595878, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.7994686365127563, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/789efc6c7a9686676edac7c87b3af6a2e378d0812ee281fbbb65b550fba6e49f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "78a39636efcb434c7c9135cd3235a7baed5f631c67483770701de076dadfc133", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98afc80f86e63e2244ad4c2c63ffd7ac3eadeece32bc5afe3d26aa309071f07f", "generation_seconds_batch": 23.71445237705484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.568817377090454, "target_z": 0.1088866493718403, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.7609207630157471, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/78a39636efcb434c7c9135cd3235a7baed5f631c67483770701de076dadfc133.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "78e25b675ab6744d78ee5e1da7b2a6f944c9ea67d56a55aa671b51e61b8b3b2d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "088f1943b6e80382341bbd1bcd0f354831842bf874c60568ff939a85b5cfcbe4", "generation_seconds_batch": 23.65204745705705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205624580383301, "target_z": -0.6562498567676703, "ce": 6.24500846862793, "genuineness": 4.356721878051758, "identity": 0.5083550214767456, "hypothesis": "ik heb het kleine doosje gevonden het stond naast het oude raam we kunnen er samen in kijken", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/78e25b675ab6744d78ee5e1da7b2a6f944c9ea67d56a55aa671b51e61b8b3b2d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "78e6fbaef587335b127bd4346d64d55d93508718b3d5ffb2419470acb4961f8e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69d2482f5ed118dfaa28eadefd3067146e6778b0d4c4a7d237e65acebab32d1f", "generation_seconds_batch": 24.120508004911244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3733454942703247, "target_z": 0.34375, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.6864384412765503, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/78e6fbaef587335b127bd4346d64d55d93508718b3d5ffb2419470acb4961f8e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "793bc2c469e6efa9f37fd94de32385e48d8611de67ee03c83834aa5dab00727a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7310dfe2a0d59a0cbb8e56ca6adff8295a2bd3571f27fb2862d96830ff8a7507", "generation_seconds_batch": 25.155860129976645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.515580654144287, "target_z": 0.17285162964015455, "ce": 6.046441555023193, "genuineness": 4.582659721374512, "identity": 0.6609264612197876, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/793bc2c469e6efa9f37fd94de32385e48d8611de67ee03c83834aa5dab00727a.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very relaxed and unstrained. 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": "17c43c1f260164bbf69b5c75530b2746ffa4cc2e776106e264829092f58a337e", "generation_seconds_batch": 18.284540520980954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8658857345581055, "target_z": -0.2480468839520206, "ce": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.7515055537223816, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "795890d486266239acd38f1531a05ddef98d15196e36bbf6ecdc926b417a0ad5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9879120570183f1ef199b2b1734457bb27623022952436d6074792bc4d6f65d6", "generation_seconds_batch": 25.597969933995046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6838246583938599, "target_z": -0.029296883952020607, "ce": 5.95112943649292, "genuineness": 4.833701133728027, "identity": 0.6919798851013184, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/795890d486266239acd38f1531a05ddef98d15196e36bbf6ecdc926b417a0ad5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "7a0fc81256779590791646b226d2c0ff0136aa32a6b80971d19ea269e44aebd8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d9343cad632d76a7a5dea4a9cacce54eec8cb9357d0e4ca3315dd44860e94a4", "generation_seconds_batch": 24.99434576299973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.199122428894043, "target_z": -0.6484373925757527, "ce": 6.0305562019348145, "genuineness": 4.469690799713135, "identity": 0.7098666429519653, "hypothesis": "De deur staan nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7a0fc81256779590791646b226d2c0ff0136aa32a6b80971d19ea269e44aebd8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "7a23bde225b0605880d812cb563cfd7adc7b8b74994e778293d3fce910881195", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b4e0d7957095d2cceb69b4992a7f464798afd9ba22ed33b3afc1dc222447eac", "generation_seconds_batch": 19.69857766502537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8455663919448853, "target_z": -0.2236327901199485, "ce": 6.046441555023193, "genuineness": 4.796044826507568, "identity": 0.7380874752998352, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7a23bde225b0605880d812cb563cfd7adc7b8b74994e778293d3fce910881195.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "7a579b853bc3722787057abd746872abe115ce6bbb7ccc763d7ad8cc1c7533a6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54454170142be40c917d73e3e826180205507b90357463b210fda9beb5649135", "generation_seconds_batch": 18.88197270198725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6769161224365234, "target_z": -0.020996140748108186, "ce": 6.252951145172119, "genuineness": 4.645420074462891, "identity": 0.5876948833465576, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7a579b853bc3722787057abd746872abe115ce6bbb7ccc763d7ad8cc1c7533a6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "7ae03c7f630ae6fa366c4d803a0d439ca5b2374eea0a65fd24dd139759962bf0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3718cdf8742e17135a749dd3ee71ba9ee17bffe082a41b72a7c5665ecac47e60", "generation_seconds_batch": 26.839735769899562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8016767501831055, "target_z": -0.17089851359217517, "ce": 5.974957466125488, "genuineness": 4.281409740447998, "identity": 0.7607388496398926, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7ae03c7f630ae6fa366c4d803a0d439ca5b2374eea0a65fd24dd139759962bf0.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": "7ae5fcc4a16fbe67fe3b2344f15cececa9a4647822ea20beb4fe50d4810e55ce", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e31740d7d8226b70294889f7d0850d4b161fbabe0f872c1554a7dbfd9decf6a6", "generation_seconds_batch": 19.48033198202029, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8098044395446777, "target_z": -0.18066409383207213, "ce": 5.919358730316162, "genuineness": 4.8588056564331055, "identity": 0.7330222129821777, "hypothesis": "Tu mensaje llegó después de que me fuera, lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7ae5fcc4a16fbe67fe3b2344f15cececa9a4647822ea20beb4fe50d4810e55ce.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "7b4b92ab89ab552d28aae5e07dbd0a79b2bbde1686860dac6b93e61de71cb0a2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "bf38ce2e58812e492d524b203568cc77784704ffd877bcabc495bb985cb45b93", "generation_seconds_batch": 18.16572177200578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5907621383666992, "target_z": 0.0825195827241185, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.8187676668167114, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7b4b92ab89ab552d28aae5e07dbd0a79b2bbde1686860dac6b93e61de71cb0a2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "7cc3ea3c6f961a94947683780dcfc22086222c9eba37fa440d72a20bf2c0a099", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c7b653c7688c360dcaaef7fad08455d5cc2f88830dd8c1675b719072847dde22", "generation_seconds_batch": 19.422527996939607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.887017846107483, "target_z": -0.27343753580808244, "ce": 6.284721851348877, "genuineness": 4.105680465698242, "identity": 0.5254546999931335, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7cc3ea3c6f961a94947683780dcfc22086222c9eba37fa440d72a20bf2c0a099.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "7cdb498a0f4f0c988f1e8782665b2f20eeb84396f3f694f706eb0e17f0202899", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1dd4ff123f80dea34e44f97f31f09b927597e60eed65284dbf8ef0f6742377a6", "generation_seconds_batch": 17.13349960197229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4529972076416016, "target_z": 0.2480468839520206, "ce": 6.213237762451172, "genuineness": 4.570107460021973, "identity": 0.6367691159248352, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7cdb498a0f4f0c988f1e8782665b2f20eeb84396f3f694f706eb0e17f0202899.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "7d76627eb04d4a1ea9edd02c05b1f91a82fef8f2ca641a0c2a31d9d2097cccde", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9db04299d16ebf9e348911c89459fab1bc82049296dd2a3ddce574b34781a8ba", "generation_seconds_batch": 24.71491583099123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6513137817382812, "target_z": 0.009765580239896965, "ce": 6.173524379730225, "genuineness": 4.8588056564331055, "identity": 0.8060816526412964, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/7d76627eb04d4a1ea9edd02c05b1f91a82fef8f2ca641a0c2a31d9d2097cccde.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "7e3a2b41660f357805aefdd637cacea7ba2fa1dfc4ee2c1a1e66924c391864d5", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "b354d410ea05a7bb47af071f472aaa00857f6eaebd16305526cd1e31a958d98c", "generation_seconds_batch": 16.803423152072355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2920682430267334, "target_z": 0.4414062320959588, "ce": 6.125868320465088, "genuineness": 4.68307638168335, "identity": 0.5947600603103638, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7e3a2b41660f357805aefdd637cacea7ba2fa1dfc4ee2c1a1e66924c391864d5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "7e76c6136a645d1a67cd1231ce4910700b1e43a30e2842133b87335dcd9d40c6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06163add0e38c52b2fd165519cb93b20ec69cfa18150fd85f5002120a15db0f7", "generation_seconds_batch": 24.537829334032722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127598524093628, "target_z": -0.5625, "ce": 5.935244083404541, "genuineness": 4.174716949462891, "identity": 0.5286087989807129, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7e76c6136a645d1a67cd1231ce4910700b1e43a30e2842133b87335dcd9d40c6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "7f173a9cceced8911ab6e5a03193ebaaef73b3d6d4022d4f078a84a2ba1937f5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "77e940521a0ee669e3f413fc2e0c5b8bf7120a4d61ddc828e720bac61e582778", "generation_seconds_batch": 22.878705587936565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4464950561523438, "target_z": 0.2558593481439382, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.48788270354270935, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7f173a9cceced8911ab6e5a03193ebaaef73b3d6d4022d4f078a84a2ba1937f5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "7f2c8dc8a08e3ad26eeb5a5ad5fcf79816e7b4c000bab73fba520f0b6ccb9e8c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c8f22be2a6503b1bf96954b9ab5688208a365595a7e2619c3d398f8488534eb", "generation_seconds_batch": 24.79854503890965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6984546184539795, "target_z": -0.04687507161616486, "ce": 6.157639026641846, "genuineness": 4.494794845581055, "identity": 0.41222935914993286, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7f2c8dc8a08e3ad26eeb5a5ad5fcf79816e7b4c000bab73fba520f0b6ccb9e8c.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": "7f650c1c8722ed2c3bd554694a8abbbbe2e5bfb00ac7a4fe8d74670731059844", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58cb3b4799ee599484dab547d6b14383db4d63bcedaa0cf422dbabe5ad621425", "generation_seconds_batch": 24.882101555005647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4481205940246582, "target_z": 0.2539062320959588, "ce": 6.300607204437256, "genuineness": 4.6956281661987305, "identity": 0.7080784440040588, "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_TENS_reg_low/7f650c1c8722ed2c3bd554694a8abbbbe2e5bfb00ac7a4fe8d74670731059844.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very relaxed and unstrained. 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": "5b345ba1c12240e952ca564c99abb3569de4b3afb135cb7687effad819dcaa66", "generation_seconds_batch": 18.010320142027922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.878890037536621, "target_z": -0.26367181233585574, "ce": 6.395919322967529, "genuineness": 4.281409740447998, "identity": 0.504584789276123, "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_TENS_reg_low/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "7ff0db57c69d2405fe8861465b825bb1d773a05ce0472c30fc84c0ffe053d540", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a42426bd5074d73d5912dbf1d064bd946bfd6d153c900c35ceab84e92eb85d8", "generation_seconds_batch": 17.85253371600993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9812994003295898, "target_z": -0.38671869628787636, "ce": 5.8875885009765625, "genuineness": 4.13078498840332, "identity": 0.6461713314056396, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/7ff0db57c69d2405fe8861465b825bb1d773a05ce0472c30fc84c0ffe053d540.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8018fe9eb306631cec420374b671da3c1d8cd89ef6c9fab9702d1f3a2d92c4b1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e6213e47251f3e627974332ee56ca704ddeabd2f8a778549542a66da3fcd76b", "generation_seconds_batch": 25.566503594978712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5594704151153564, "target_z": 0.12011720988005152, "ce": 6.38003396987915, "genuineness": 4.432034492492676, "identity": 0.6366814374923706, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschaffen.", "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_TENS_reg_low/8018fe9eb306631cec420374b671da3c1d8cd89ef6c9fab9702d1f3a2d92c4b1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "803e2a4028c638137f2dbc4d181a7882557473045db98e5cb77ca2dd678ca825", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e0b5a474b277df177f447e0d3abcb3c0976fe34d75a48f5f28ee39db8b1f818", "generation_seconds_batch": 25.95990879496094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.604172945022583, "target_z": 0.06640623209595879, "ce": 5.967014789581299, "genuineness": 4.482243061065674, "identity": 0.6172481179237366, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/803e2a4028c638137f2dbc4d181a7882557473045db98e5cb77ca2dd678ca825.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "804e65db352414dc9fe7d0e4e722902766ffca806433e3b7f1cb8b863745392f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af4c9300425fe8d4b143c01e75a0a5fe856e73f9681f6d790d52cc87e2e0f6fc", "generation_seconds_batch": 26.47588178003207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.420486330986023, "target_z": 0.2871093481439382, "ce": 5.677107334136963, "genuineness": 4.218649387359619, "identity": 0.6391081809997559, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/804e65db352414dc9fe7d0e4e722902766ffca806433e3b7f1cb8b863745392f.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": "8061994221445bc143772c41f0a7e116ca78c1fdd1e46fe361fb929440aade17", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3338e67e2e4aee5d92a263483a0167a050831f86edc29be1d5c78dfbe0bfa0a4", "generation_seconds_batch": 27.74185279000085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5407766103744507, "target_z": 0.14257818766414426, "ce": 5.935244083404541, "genuineness": 4.557555198669434, "identity": 0.6767303943634033, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8061994221445bc143772c41f0a7e116ca78c1fdd1e46fe361fb929440aade17.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "806ce229d87b142ca06e84bc11c5228ad81ef5576e44fe4bda348dbb7c478a74", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ba1202d4342699e0fa432a84a11c3af801b7e18c0fa604d78c7ecf681bddfe1d", "generation_seconds_batch": 23.66779607301578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5131423473358154, "target_z": 0.17578130371212364, "ce": 6.165581703186035, "genuineness": 4.632867813110352, "identity": 0.683735728263855, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/806ce229d87b142ca06e84bc11c5228ad81ef5576e44fe4bda348dbb7c478a74.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "80bb2a6ca0d6761c69042c6d9a348ae18320f35bd720409c7b003a7cfac0e1d8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c21628869785a3cf069a99c5eee93bdb77ce5ebcbe178b0ec21bf686c4358383", "generation_seconds_batch": 22.92645551392343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6106750965118408, "target_z": 0.058593767904041215, "ce": 6.332377910614014, "genuineness": 4.419482707977295, "identity": 0.24697786569595337, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/80bb2a6ca0d6761c69042c6d9a348ae18320f35bd720409c7b003a7cfac0e1d8.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very relaxed and unstrained. 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": "989da3f4e7e356dedaaaf582bd9ae0e63791c8ffed9d46ded2800b6e28ce0da6", "generation_seconds_batch": 19.72736270702444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7699785232543945, "target_z": -0.13281246419191758, "ce": 6.23706579208374, "genuineness": 4.432034492492676, "identity": 0.6297292113304138, "hypothesis": "Las llaves están en la mesa de la cocina. Mire en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "80fd5e287fecf2b69a75a51372729e9b038ead602283de5a3b2d1aae79bcf98a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88f16ce51d0b6dbd6d87b995e065517bde4a54e226b0d847536b4ecc43b3bf53", "generation_seconds_batch": 23.479900148930028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4318650960922241, "target_z": 0.27343753580808244, "ce": 6.070269584655762, "genuineness": 4.055472373962402, "identity": 0.6580253839492798, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/80fd5e287fecf2b69a75a51372729e9b038ead602283de5a3b2d1aae79bcf98a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "810119fe9c04876505803bb51016a281506a105c827e66903451b19b3d3c30d0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "08999095ab4cc872cc820af3244056fd0a095a3ab394f929ce301c57b3da83f5", "generation_seconds_batch": 25.643444009008817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6171772480010986, "target_z": 0.050781303712123645, "ce": 6.252951145172119, "genuineness": 4.469690799713135, "identity": 0.6618756055831909, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfahren.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/810119fe9c04876505803bb51016a281506a105c827e66903451b19b3d3c30d0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "811e217665acf4d733b7b9e9341cb3aa8ef44c9c219fe8a372a960de945f5ba2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14b910967edaadd50f8773044b9ac6b946ac87e13b3dff501e1df8bf72beb1d2", "generation_seconds_batch": 22.49123009899631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0690789222717285, "target_z": -0.49218753580808244, "ce": 6.181467056274414, "genuineness": 4.400654315948486, "identity": 0.5531973838806152, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/811e217665acf4d733b7b9e9341cb3aa8ef44c9c219fe8a372a960de945f5ba2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "81251077a3f6ff00f3c180243de3e195c766a62bc41293a1526c5ea561bb107e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ecc60f88840062979e4e73ba04fbc23c383c030be01333f27101fc93a19eb1b2", "generation_seconds_batch": 22.578386905952357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1763648986816406, "target_z": -0.6210937679040412, "ce": 6.173524379730225, "genuineness": 4.369274139404297, "identity": 0.7170917391777039, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/81251077a3f6ff00f3c180243de3e195c766a62bc41293a1526c5ea561bb107e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8135e2553eca1daf24414fcbac43dfa9a8136fa536cf4c65e3ce35b1fdff8938", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3540a2095743f2eba8942815ee8f1a4a9290e195ce9d6459923d18ebce361b95", "generation_seconds_batch": 24.5956024780171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7968000173568726, "target_z": -0.16503902221590727, "ce": 6.427689552307129, "genuineness": 4.494794845581055, "identity": 0.7602502107620239, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8135e2553eca1daf24414fcbac43dfa9a8136fa536cf4c65e3ce35b1fdff8938.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": "816857f0cd595204a7e958e583498af969164327ef9ddc6f4629cd9bfd4b8f55", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c38dc3a6740bf2a60f7585af200af4aeaa902abda76c30b25d23dfe1b1839e22", "generation_seconds_batch": 15.864845247939229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8049278259277344, "target_z": -0.17480474568813395, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.3807485103607178, "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_TENS_reg_low/816857f0cd595204a7e958e583498af969164327ef9ddc6f4629cd9bfd4b8f55.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "81af554f98b0f776f66a90148ff071f03801474dea0a7ddd83b85deccf851392", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4acf7b68c6c82bb5004be6fff43b06d435ad10b391dd553311cadc60fb168dd0", "generation_seconds_batch": 22.82235076301731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7187738418579102, "target_z": -0.07128902221590727, "ce": 6.435632705688477, "genuineness": 4.306513786315918, "identity": 0.5668401718139648, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/81af554f98b0f776f66a90148ff071f03801474dea0a7ddd83b85deccf851392.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "81b5f958d16f1f1cc5b871388bc40edc3f0d729bbf873fae1dd7c512868eff89", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9eb15e7417977f3eb4850acb248034edf14880c6d4d7cc6404a9f97eb2031be0", "generation_seconds_batch": 24.642908500041813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.308323621749878, "target_z": 0.42187507161616483, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.7175976037979126, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/81b5f958d16f1f1cc5b871388bc40edc3f0d729bbf873fae1dd7c512868eff89.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "821424441f5a2c673c8d7d95d66169344f958509f61d11f740387eb85bef34d4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14f88420c8f09651a9b7cd33ba278de2a3724a56bb840a8076fd0ac2a4ae145a", "generation_seconds_batch": 19.062023898935877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7862340211868286, "target_z": -0.15234376790404122, "ce": 6.308549880981445, "genuineness": 4.482243061065674, "identity": 0.3190140128135681, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/821424441f5a2c673c8d7d95d66169344f958509f61d11f740387eb85bef34d4.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": "821eeaac93024ddd46911cedc561d0dcd52746263c25cc66023a1427b62f4f1d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1e9b902141963594dc56dfc59852be24b51ef3edc51b00438d65f5c674d3b04", "generation_seconds_batch": 20.47187789203599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0625767707824707, "target_z": -0.48437507161616483, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.7052359580993652, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/821eeaac93024ddd46911cedc561d0dcd52746263c25cc66023a1427b62f4f1d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "822b0cad9858d718d0ebf6e23adebc98ad4f7095c80f56ee5e2920cca502881a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ac57817020572cf2a69590903a30ae4fc71553707b86333880f0eca97057dfd", "generation_seconds_batch": 23.65204745705705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.312910556793213, "target_z": -0.7851560888636291, "ce": 6.300607204437256, "genuineness": 4.557555198669434, "identity": 0.44373247027397156, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/822b0cad9858d718d0ebf6e23adebc98ad4f7095c80f56ee5e2920cca502881a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "826b2f171c9bd6bc948705c54b8348354b4fe0d6023b168afcc74a437a00bf25", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d71eb19c028d2fcb40ee70dd866abe1895842ae096a7361f2aae99f5e71bbe2d", "generation_seconds_batch": 16.565960846026428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7386868000030518, "target_z": -0.09521483703598455, "ce": 6.1894097328186035, "genuineness": 4.287685871124268, "identity": 0.5224032998085022, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/826b2f171c9bd6bc948705c54b8348354b4fe0d6023b168afcc74a437a00bf25.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "82bff8bc22f5fac37215503724d11c68bacb3f01fba843463a75816e07b577ec", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d74ba718c7b0b5c50984d0918de514d0ea60a14f47c731016a09df463dd3ebce", "generation_seconds_batch": 25.090283766039647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7183674573898315, "target_z": -0.07080074320391241, "ce": 5.7406487464904785, "genuineness": 4.149612903594971, "identity": 0.4915338456630707, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/82bff8bc22f5fac37215503724d11c68bacb3f01fba843463a75816e07b577ec.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "82d42e28fc48ddc4eb5353f0b46019f15b76869b8442731f78f7adbbcec9173b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a77afdd02998d734a40997cd790bc1f64a44b22b4dccd4526d5fcd93f36b7e2", "generation_seconds_batch": 25.664533686009236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7976127862930298, "target_z": -0.16601558023989696, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.47102296352386475, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/82d42e28fc48ddc4eb5353f0b46019f15b76869b8442731f78f7adbbcec9173b.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": "8395585ddf68f635d9d8e136922622c260070ac3f68313870e741d14e00af5c8", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "2fdf828928cc21e517dbe765c4644710667ef3bf910e04708875daec99aef3a7", "generation_seconds_batch": 20.291013185982592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7968000173568726, "target_z": -0.16503902221590727, "ce": 5.95112943649292, "genuineness": 4.507347106933594, "identity": 0.6971398591995239, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8395585ddf68f635d9d8e136922622c260070ac3f68313870e741d14e00af5c8.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": "83d3ff4a28ed3fe71a8e55c2e9787530871e095f13ed84482d0f3f2d36c4e226", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90925917c873ae71ad14d7bb6508853951a6313afb9ca851a65b134247a23983", "generation_seconds_batch": 19.858084617066197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.593606948852539, "target_z": 0.07910148640782484, "ce": 6.165581703186035, "genuineness": 4.494794845581055, "identity": 0.5609045028686523, "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_TENS_reg_low/83d3ff4a28ed3fe71a8e55c2e9787530871e095f13ed84482d0f3f2d36c4e226.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "8434114e8ccaa202e3cc117a29d60a14ce44c387af426c43ff23f6a97a464ad1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d4a5be52497a00d521cb4119d80eea4f52344dba5ffc839e4d3cceb79a578e9", "generation_seconds_batch": 23.436586808995344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4334906339645386, "target_z": 0.271484419760103, "ce": 6.308549880981445, "genuineness": 4.293961524963379, "identity": 0.3444821238517761, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8434114e8ccaa202e3cc117a29d60a14ce44c387af426c43ff23f6a97a464ad1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "84a4ed0c2d0bd615390554b7b7eca0691f014c863f886bdc806251eb6cd09d19", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7b11d40f3fa769b8ef21e1f803e9af8eab8649d9bab096d8cb82246e381e6207", "generation_seconds_batch": 15.243745477986522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.23706579208374, "genuineness": 4.331617832183838, "identity": 0.18931138515472412, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/84a4ed0c2d0bd615390554b7b7eca0691f014c863f886bdc806251eb6cd09d19.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "84b91dd5558ec057a65f067cdcbc75b6f41eb2442ccb19fa86a722977d38d101", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1dc16fb1f1401dd022a36e2a9cdcf964a43074b1e1699dd5a80135a9ff6c90e", "generation_seconds_batch": 23.42114908702206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2010376453399658, "target_z": 0.5507813037121236, "ce": 5.609594821929932, "genuineness": 4.18099308013916, "identity": 0.733107328414917, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/84b91dd5558ec057a65f067cdcbc75b6f41eb2442ccb19fa86a722977d38d101.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "84c796bc15f76fa3d205b0459cdabf171e4e34c7e489f26cedfea507cb623720", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a79b85e5418423fb0ea72ef936b51673593e7758cfd98f787226378f4b734c86", "generation_seconds_batch": 18.06196842691861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6591366529464722, "target_z": 0.00036624506707856467, "ce": 6.348263263702393, "genuineness": 4.300237655639648, "identity": 0.260951429605484, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/84c796bc15f76fa3d205b0459cdabf171e4e34c7e489f26cedfea507cb623720.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": "85fd4a6462973ffada4b68603d5985daa769d7344fbd06a0f2fd13105795667f", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "118f55dfa7203422ad7593a7612707e16cb8c724e05d1406f9ae105be55a41e1", "generation_seconds_batch": 23.830914153018966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5342744588851929, "target_z": 0.15039065185606182, "ce": 6.006728172302246, "genuineness": 4.507347106933594, "identity": 0.8206073641777039, "hypothesis": "La route tourne à gauche après le pont continue jusqu'à la petite barrière en bois j't'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/85fd4a6462973ffada4b68603d5985daa769d7344fbd06a0f2fd13105795667f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "8604872d098b4ac016b62e7cab1d3c3056f9824d92f20e8d01b610a5d33512ee", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3756e6ab9c880d6e0a3698c231e1d87e47a15a509fb70636439c401bfcf5fab", "generation_seconds_batch": 24.517615389893763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.156858444213867, "target_z": -0.5976563753282885, "ce": 5.824047088623047, "genuineness": 3.9425034523010254, "identity": 0.29513245820999146, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8604872d098b4ac016b62e7cab1d3c3056f9824d92f20e8d01b610a5d33512ee.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "8616e9f816b0f4f3ca7d5b9f5798422c6be5a7561522a36d65c374b979e0294c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "104696454697d099be2343cef60098b28430ae7411cea2c56ff9e8c02abc3cfc", "generation_seconds_batch": 24.933715575025417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8723878860473633, "target_z": -0.2558593481439382, "ce": 6.125868320465088, "genuineness": 4.570107460021973, "identity": 0.6974759101867676, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8616e9f816b0f4f3ca7d5b9f5798422c6be5a7561522a36d65c374b979e0294c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "8625286cc412beadd1372dce6f9770aa8123746b274f4319c62ab837d181a2e2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6de061c3139ad5c7fa988fe590fe7c1f8071c1e3167cb97fb24f74268f1c70d", "generation_seconds_batch": 19.338564528967254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6635053157806396, "target_z": -0.004882790119948482, "ce": 5.80021858215332, "genuineness": 3.8169827461242676, "identity": 0.6643434166908264, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8625286cc412beadd1372dce6f9770aa8123746b274f4319c62ab837d181a2e2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "8641afdef953f7e62e652e235926766a9d9b1d8f3cc29005f6ebc73478ef602f", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "91d6728615b6a65a2adaff6c59dd445ed7dadd05175ca824654bfc70af965c73", "generation_seconds_batch": 20.307066318928264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.280399799346924, "target_z": -0.7460937679040412, "ce": 6.332377910614014, "genuineness": 4.557555198669434, "identity": 0.7346682548522949, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8641afdef953f7e62e652e235926766a9d9b1d8f3cc29005f6ebc73478ef602f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "86b2de762ae8c31bf7e2a036aad4b047e3fbe2ed088517fa65c0a09726a4cc00", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6f502762c02d119868e34f83362e7601c7e72011d14517aa739e04b56f18072", "generation_seconds_batch": 24.037384828086942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.085334300994873, "target_z": -0.5117186962878764, "ce": 6.300607204437256, "genuineness": 4.193545341491699, "identity": 0.41299837827682495, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/86b2de762ae8c31bf7e2a036aad4b047e3fbe2ed088517fa65c0a09726a4cc00.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": "86eef4804c8d6fe97d1dbb9d60f87119b00cefac8cb7912cc0c61df2174fbdb5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "ee477bf7526cf477790490c8938a7bfddd73d60b1dc2a98f7e665a5002f35427", "generation_seconds_batch": 18.900832889950834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5131423473358154, "target_z": 0.17578130371212364, "ce": 5.9431867599487305, "genuineness": 4.632867813110352, "identity": 0.6895779967308044, "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_TENS_reg_low/86eef4804c8d6fe97d1dbb9d60f87119b00cefac8cb7912cc0c61df2174fbdb5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "86f810fd81227ef54c8358e79a228a8c45a958bf0d18cd8e48aea1657a8b17a7", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0b2ec1ded362f2ad594cac0e7f5492a8390fa619a98e49519df4723e5f39af17", "generation_seconds_batch": 19.180992110050283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6192091703414917, "target_z": 0.048339908652149405, "ce": 6.24500846862793, "genuineness": 4.68307638168335, "identity": 0.8188703060150146, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/86f810fd81227ef54c8358e79a228a8c45a958bf0d18cd8e48aea1657a8b17a7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "87025b82d6bebf478a8bdfd48a5ec333cad8d181fd41301c9cabd74cb9080958", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eb195893c3255c0adb9be7ec743f8cfdb53d2a67ebaa8b1c2a1346567f8eea9b", "generation_seconds_batch": 26.549365081940778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1490201950073242, "target_z": 0.6132813037121236, "ce": 5.903473377227783, "genuineness": 4.557555198669434, "identity": 0.8036051988601685, "hypothesis": "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_TENS_reg_low/87025b82d6bebf478a8bdfd48a5ec333cad8d181fd41301c9cabd74cb9080958.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8758cd0bdccad3e643069b124482a41c7fb55fc9b9d499636c7ff40ab0176295", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c60c22b51610ddfc4ab7e9fd99a152a4a07b4c855d3d563f02a37d21e42d9cd", "generation_seconds_batch": 24.701112549984828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5424022674560547, "target_z": 0.14062492838383514, "ce": 6.332377910614014, "genuineness": 4.218649387359619, "identity": 0.5830111503601074, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8758cd0bdccad3e643069b124482a41c7fb55fc9b9d499636c7ff40ab0176295.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "87ce2c8b401215da2c1d6ce058ae96f4c5abfd704bcff4b4cac6429582e960be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d2d845d017d5cd89252845e1eb77baa20c8edbd01d2a42afd109a006875e28e", "generation_seconds_batch": 25.03373357397504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2042887210845947, "target_z": 0.5468750716161649, "ce": 5.669164657592773, "genuineness": 4.632867813110352, "identity": 0.7099809050559998, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/87ce2c8b401215da2c1d6ce058ae96f4c5abfd704bcff4b4cac6429582e960be.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "87d467f8a5abbb8bfccf638283b1fd6eafe235de0b498bd0998a71ec995b8aed", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1ff3b9619b6cf28586d5528c2aef5ecc59d51630d140a6a525560d138d7bb7a7", "generation_seconds_batch": 17.31516316498164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2218801975250244, "target_z": -0.6757813037121236, "ce": 5.879645347595215, "genuineness": 4.457139015197754, "identity": 0.5806350708007812, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/87d467f8a5abbb8bfccf638283b1fd6eafe235de0b498bd0998a71ec995b8aed.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "87eec8b8648318affc1d67242d6ffd1edbc0e027b259f22fcf38cc534c81ee8a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65bb50b2d6e597c7f99737e050584e958d49990b9f4a534bf928e55c6a61c4c1", "generation_seconds_batch": 23.064952155924402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.034942388534546, "target_z": -0.45117181233585574, "ce": 6.308549880981445, "genuineness": 4.293961524963379, "identity": 0.5852352380752563, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/87eec8b8648318affc1d67242d6ffd1edbc0e027b259f22fcf38cc534c81ee8a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "882ae8a9ab035c21eb3eb0429d5177b2f9e834d9eeed089c5090af51a84a0d83", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd75e98ab4cbe425f49b0f6cb1d76b0843323c6b8a75a7d88123d542002220f8", "generation_seconds_batch": 23.50184205302503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127598524093628, "target_z": -0.5625, "ce": 6.332377910614014, "genuineness": 4.607763767242432, "identity": 0.7019374370574951, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/882ae8a9ab035c21eb3eb0429d5177b2f9e834d9eeed089c5090af51a84a0d83.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "884b2038a3657cdb30125d730f9e58012d39495a716bbac825b34f79d675c970", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75821c4318a07f88e38c0ba59d4e42d08f3d03fe874157ebe61214a88fba6cd7", "generation_seconds_batch": 15.658181454055011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9422863721847534, "target_z": -0.3398437679040412, "ce": 6.24500846862793, "genuineness": 3.9425034523010254, "identity": 0.21915626525878906, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/884b2038a3657cdb30125d730f9e58012d39495a716bbac825b34f79d675c970.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": "88612399e3a14d4bc867557fd0ee38739743e5da0c26c39cd10dccd4d61e640e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "15267a1bb63f131057682dccd9d40431c1891f10d0532c7bc13c56c338ae9067", "generation_seconds_batch": 23.91047299990896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3944776058197021, "target_z": 0.3183593481439382, "ce": 6.046441555023193, "genuineness": 4.65797233581543, "identity": 0.7095556259155273, "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_TENS_reg_low/88612399e3a14d4bc867557fd0ee38739743e5da0c26c39cd10dccd4d61e640e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "8892caa82e60e319470b94fee88a7f44f9f1b2e779151ab34c23fa77d230e1eb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f62c4db143337fb0d6b8cb6084ad3ed4b6335183f4265b47379cf7c1282dcbb9", "generation_seconds_batch": 23.769813135964796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.75372314453125, "target_z": -0.11328130371212364, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.7337104082107544, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8892caa82e60e319470b94fee88a7f44f9f1b2e779151ab34c23fa77d230e1eb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "89098a6d9a750cce05d71b32f5ec95811d71e16cedf5c0e3ee7e74b759f838cc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4ef462884bbea7d9e5493bed4dffd75a7e742269a3d37224f4f74ab7c1f72a69", "generation_seconds_batch": 24.201527010067366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3180769681930542, "target_z": 0.4101562320959588, "ce": 5.911416053771973, "genuineness": 4.137060642242432, "identity": 0.3841521739959717, "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_TENS_reg_low/89098a6d9a750cce05d71b32f5ec95811d71e16cedf5c0e3ee7e74b759f838cc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8938749ac6595ee8985cb5258b11c3be60c8aeb6dc9546e9cea11ddcd50825a0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff1d58de6582475c1fe7351efb9f86f4bfcbe757c82a3800ba25567de5ffbe39", "generation_seconds_batch": 23.76468881999608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.714709997177124, "target_z": -0.06640623209595879, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.6323891878128052, "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_TENS_reg_low/8938749ac6595ee8985cb5258b11c3be60c8aeb6dc9546e9cea11ddcd50825a0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "896885ca77c1c9247dd4c3457f3c7c549b28aebb8911ae051d545294b2ef2455", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8ab44ef6bbec8b841af17c10d2a68e31a6753a0912754b264b82024b5d8a5641", "generation_seconds_batch": 24.5956024780171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.61819326877594, "target_z": 0.04956053456597167, "ce": 6.38003396987915, "genuineness": 4.425758361816406, "identity": 0.7103958129882812, "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_TENS_reg_low/896885ca77c1c9247dd4c3457f3c7c549b28aebb8911ae051d545294b2ef2455.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": "8972b8f98c3d1ba5a3a83734346c3a41aa2951dca1593729922643220ff5aceb", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "44cd1616d5999452fea1994c86480546c56bf4cad19e0cd31f0ca84751105307", "generation_seconds_batch": 20.07211304199882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9991804361343384, "target_z": -0.4082031160479794, "ce": 6.09409761428833, "genuineness": 4.570107460021973, "identity": 0.7562901973724365, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8972b8f98c3d1ba5a3a83734346c3a41aa2951dca1593729922643220ff5aceb.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": "89b53723b2f0411ad00e924f3a8e2e9ee6ac175153362a1e158281e00ad0dc7c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "098a60703deb478dabb34ec14da23d8be1303c1a5b28c5aae87fb1c79d107567", "generation_seconds_batch": 25.25871634494979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7017056941986084, "target_z": -0.050781303712123645, "ce": 6.284721851348877, "genuineness": 4.337893962860107, "identity": 0.32029345631599426, "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_TENS_reg_low/89b53723b2f0411ad00e924f3a8e2e9ee6ac175153362a1e158281e00ad0dc7c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "89c6bf1739afc360f9e5235598e4de9bef360035da8430062df0aff15a6a357c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7350420fcc4fa17ddf5d6a34b62b6c4569bc69c4ac7818794f724797f18884fa", "generation_seconds_batch": 18.620863603893667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.953954815864563, "target_z": 0.8476562320959588, "ce": 5.95112943649292, "genuineness": 4.545003414154053, "identity": 0.7032431364059448, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/89c6bf1739afc360f9e5235598e4de9bef360035da8430062df0aff15a6a357c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "8a24ecea7ae218965b7fb349f31959990229a13bacb35ddaf9597e9e07860eac", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0c101dd95d2a3fcaed3bfd210eb0207208599edc4350f1b83bfd7028f68fa3ba", "generation_seconds_batch": 19.29479461198207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.637699842453003, "target_z": 0.026122998757889235, "ce": 6.292664527893066, "genuineness": 4.3127899169921875, "identity": 0.3659367561340332, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8a24ecea7ae218965b7fb349f31959990229a13bacb35ddaf9597e9e07860eac.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "8a8239c8fd12c2f0863c37b680256d0c765ada73478bdd027abbdec5a1067eeb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd4c5d0d461ee9af909e2f3ba8fed3883ce31e8bc88853000a8145e96790d28b", "generation_seconds_batch": 25.53537341102492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1847822666168213, "target_z": 0.5703124641919176, "ce": 5.816103935241699, "genuineness": 4.595211505889893, "identity": 0.7598851323127747, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8a8239c8fd12c2f0863c37b680256d0c765ada73478bdd027abbdec5a1067eeb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "8aa0661af0c9bdf91b82b5e6fdefbd32f9c4c3a183ee5efca54350b2eb5495e6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3435e1ac1a3104f3731dc417d877af892d0198417b5eb6cace0b9ef69eb7101f", "generation_seconds_batch": 18.12475952494424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7699785232543945, "target_z": -0.13281246419191758, "ce": 6.1020402908325195, "genuineness": 4.243753433227539, "identity": 0.39500418305397034, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8aa0661af0c9bdf91b82b5e6fdefbd32f9c4c3a183ee5efca54350b2eb5495e6.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": "8af2ee91437f01e19eb6d60bf8ac01447e6f61e9a010153a285e960d3f58002d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68cdf148120b1e7d6fd6db67332626934fc496b5f0e8403eaa93628da1a0435b", "generation_seconds_batch": 24.404016270069405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7598189115524292, "target_z": -0.12060548889204636, "ce": 5.9431867599487305, "genuineness": 4.557555198669434, "identity": 0.5849384069442749, "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_TENS_reg_low/8af2ee91437f01e19eb6d60bf8ac01447e6f61e9a010153a285e960d3f58002d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "8b11adbcce1940a0c5cb207bfdc2e20e659a6df1f78e57acd9377fa3f99dd86f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8333315aab21e2c3f19ede20c15a56f409489554ec6592d67d085e5ede496c8f", "generation_seconds_batch": 24.517615389893763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6425764560699463, "target_z": 0.020263650613951058, "ce": 5.824047088623047, "genuineness": 4.419482707977295, "identity": 0.3783458173274994, "hypothesis": "De deur 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_TENS_reg_low/8b11adbcce1940a0c5cb207bfdc2e20e659a6df1f78e57acd9377fa3f99dd86f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8b7e64a7e091e1913236854691f3f3e8ecfbe42363ab181718dd2e53a60a004f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "16ac214b16326874de1d9a5d6a6b88be8d5115f59e7831f4319eddb9c60cfdc0", "generation_seconds_batch": 18.32406831299886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6120975017547607, "target_z": 0.05688471974589439, "ce": 6.4515180587768555, "genuineness": 4.243753433227539, "identity": 0.6259744763374329, "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_TENS_reg_low/8b7e64a7e091e1913236854691f3f3e8ecfbe42363ab181718dd2e53a60a004f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "8be8ef422edf2a644b80130c9f1a820d3cf150cda24945d266d212aa1c03fc6a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b01a3673476d7f4e6b5674170f326acd66e39460b71b75ac21c6cfdb8ee30a0", "generation_seconds_batch": 16.810488336952403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6693471670150757, "target_z": -0.011901890437580496, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.5309914350509644, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8be8ef422edf2a644b80130c9f1a820d3cf150cda24945d266d212aa1c03fc6a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "8c1c937597310ea344fe50f1f67fa85c20ad7dbf3f4b75d9afcf6c0da831a089", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "79298b0aa370675481258fb3d0286cd11082bf2b80dede5a121bc5afcfffa20d", "generation_seconds_batch": 18.364171614055522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4286140203475952, "target_z": 0.2773437679040412, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.6632273197174072, "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_TENS_reg_low/8c1c937597310ea344fe50f1f67fa85c20ad7dbf3f4b75d9afcf6c0da831a089.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": "8c1f4a74eb47e5d9f7e1a30e23ea457c36c43bec1fb1bdadd20e421614868909", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "3eaac3080d87de171e2bc504c14281114ae32d5e225def62d8c9a5690d722dc9", "generation_seconds_batch": 21.73340437200386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4871336221694946, "target_z": 0.20703130371212364, "ce": 6.221180438995361, "genuineness": 4.7332844734191895, "identity": 0.6967188119888306, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8c1f4a74eb47e5d9f7e1a30e23ea457c36c43bec1fb1bdadd20e421614868909.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": "8c32c5f1d65b8742c670bb29a141ad8366021bc5320c479cef86ce7df4236441", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef7e0637bed855a109b1b9b9d314fbf70667677cfd735a3c38632e25ad14cee2", "generation_seconds_batch": 16.086366798030213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4806314706802368, "target_z": 0.21484376790404122, "ce": 6.038498878479004, "genuineness": 4.519899368286133, "identity": 0.704605221748352, "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_TENS_reg_low/8c32c5f1d65b8742c670bb29a141ad8366021bc5320c479cef86ce7df4236441.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "8cd59185d022dd396d956cbbcd8a89445d9084de65c164a42e2e68ea845e7a7d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38d838d2b4c053e356ae2309d456dab7c290bcc8879ba3df12562c393cc3d448", "generation_seconds_batch": 28.05661284807138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.197786569595337, "target_z": 0.5546875358080824, "ce": 6.0146708488464355, "genuineness": 4.344170093536377, "identity": 0.7065062522888184, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8cd59185d022dd396d956cbbcd8a89445d9084de65c164a42e2e68ea845e7a7d.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": "8cfe635cd7fa7c41ae4f1c882222481b457ee078fbb3b9683ed5aecabd893ad4", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "6b09e85dfb839a0c5c5c1b8e9a15fd920c4fab0f8435e28ed8e557f3cfe8d79f", "generation_seconds_batch": 21.78954419097863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7017056941986084, "target_z": -0.050781303712123645, "ce": 5.911416053771973, "genuineness": 4.7332844734191895, "identity": 0.8276015520095825, "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_TENS_reg_low/8cfe635cd7fa7c41ae4f1c882222481b457ee078fbb3b9683ed5aecabd893ad4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "8da2b2754e49e7a7ff5368b86223c01e5474913db928eb01ec9443351aede56a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7184e9aa8e04d252f6765230f787647efa5e67208bc5735398e4c31cae292e4e", "generation_seconds_batch": 18.770202643005177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7082078456878662, "target_z": -0.058593767904041215, "ce": 6.133810997009277, "genuineness": 4.243753433227539, "identity": 0.7256691455841064, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8da2b2754e49e7a7ff5368b86223c01e5474913db928eb01ec9443351aede56a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "8db00d7d58a6d7b84ce8f3e1a63294a78c36bcf084ba26597fc7981fd8a2a1e3", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7fea1f87aa4de8ef9ceeb591dc03829e23a18c5f3f83ff023b63c9096f94c237", "generation_seconds_batch": 16.250440262025222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5602831840515137, "target_z": 0.11914065185606182, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.8696494698524475, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8db00d7d58a6d7b84ce8f3e1a63294a78c36bcf084ba26597fc7981fd8a2a1e3.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": "8dc5c0018c8cc7b7ff467515ae008a1df5ac54523939e70fabfe3b9b472b974e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7de175ab7222a905c86adc87518ed2a5d835327c015e12da8f5d4ebe4324b65a", "generation_seconds_batch": 19.921026197029278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.340834617614746, "target_z": 0.38281246419191756, "ce": 6.197352409362793, "genuineness": 4.507347106933594, "identity": 0.594792366027832, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8dc5c0018c8cc7b7ff467515ae008a1df5ac54523939e70fabfe3b9b472b974e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "8e628345fdab856e04927a462320c3db24dd97da56ad41d9e582ffe8f062823a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "535722f75560c111efe132c255b5f3c4e43281fb3420a8b57b744a9a2712b77e", "generation_seconds_batch": 23.113238853984512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5440278053283691, "target_z": 0.13867181233585574, "ce": 6.0305562019348145, "genuineness": 4.519899368286133, "identity": 0.7328221797943115, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8e628345fdab856e04927a462320c3db24dd97da56ad41d9e582ffe8f062823a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "8f203f47c290d0741dea8869f2b521304aaa98a5ace4d6fe1107418ac36c6b96", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f975ce9ebb38e2d2368b12a62ca5d7f72a3c2194d62b7d4b4c2046a77d3dd4a9", "generation_seconds_batch": 18.693622352089733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0040571689605713, "target_z": -0.4140626074242473, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.6848897337913513, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8f203f47c290d0741dea8869f2b521304aaa98a5ace4d6fe1107418ac36c6b96.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "8fa0455459d7cada9584842ab8282599580cf508497504fbe4329166667e3092", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "305408364c782bc45644a18575ee384472d842f8c2b7185f2e0b043a9a8b2e48", "generation_seconds_batch": 17.43291210499592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4481205940246582, "target_z": 0.2539062320959588, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.771803081035614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what it 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_TENS_reg_low/8fa0455459d7cada9584842ab8282599580cf508497504fbe4329166667e3092.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "8fc6e94be9bf700dcb00c2d522a5c5995f3460c7c90b7b55a9331f108a7afe58", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "676e3d41f5bfcb3f7d54d30b5c54dc01de0a5e47a4fa661a1c4f7d3eaa5626dd", "generation_seconds_batch": 18.019026340916753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5952324867248535, "target_z": 0.07714837035984545, "ce": 6.252951145172119, "genuineness": 4.557555198669434, "identity": 0.48695671558380127, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8fc6e94be9bf700dcb00c2d522a5c5995f3460c7c90b7b55a9331f108a7afe58.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "8fda541c75ee41a2c0795310b68f4731d6fe6125541b35b09ec9790b4fa3eaaa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0726d5ebb60bb15e3979ff502cf62a9362cd15370268179e808f4b63595d2ec", "generation_seconds_batch": 22.770106348907575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7862340211868286, "target_z": -0.15234376790404122, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.5555191040039062, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/8fda541c75ee41a2c0795310b68f4731d6fe6125541b35b09ec9790b4fa3eaaa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "90356616bd2764f47d7ecf05c5f3b62f122275cfafb90cb0214fe12fb8fa49b1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccf087a979f513b22469e680eeb350a5ff1299896e7b30cb696166365c79572d", "generation_seconds_batch": 17.698533938033506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.701908826828003, "target_z": -0.05102537160195621, "ce": 6.197352409362793, "genuineness": 4.143336772918701, "identity": 0.4626912474632263, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/90356616bd2764f47d7ecf05c5f3b62f122275cfafb90cb0214fe12fb8fa49b1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "9039963ae3de510be2d73fd1e2ad27d597ecd058ed6866ead985172a22f3585b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92a7aa68f5f7af81e8d74000d2ff5cd5a539b38bd95eb01f68be51bf981dd5aa", "generation_seconds_batch": 22.053278245963156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6895140409469604, "target_z": -0.03613279011994848, "ce": 6.475345611572266, "genuineness": 4.758388519287109, "identity": 0.7023504972457886, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9039963ae3de510be2d73fd1e2ad27d597ecd058ed6866ead985172a22f3585b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "905e671a670b55ccbbbd8bd08fb980ac44020ff9e86c88946c2f222f196e066f", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fa76fe717ddfbfb466a2c9527f803eed5db495773d797a2ab319455f01b3e38b", "generation_seconds_batch": 16.277138358913362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6403412818908691, "target_z": 0.02294925679608758, "ce": 6.38003396987915, "genuineness": 4.325342178344727, "identity": 0.5842504501342773, "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_TENS_reg_low/905e671a670b55ccbbbd8bd08fb980ac44020ff9e86c88946c2f222f196e066f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "90667efb3e826add5e469cd3b77edeeb7d293efaf7ac43c6501040e66e624500", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10fc3c9fc75431b547131a7ee40f73b60c23d49c3a5b4c70418adcfe4b3e4a47", "generation_seconds_batch": 19.314193704980426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4976997375488281, "target_z": 0.19433590616792787, "ce": 6.24500846862793, "genuineness": 4.582659721374512, "identity": 0.7805508971214294, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/90667efb3e826add5e469cd3b77edeeb7d293efaf7ac43c6501040e66e624500.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "90a295a2d56fb83f7accf94b4a30b3cbabd1771c69c8e7572f7aab51c46e3984", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6958232b8f7e2224acc05abfc108f547770de077149a2b398a5c7c06a594005", "generation_seconds_batch": 17.504951542941853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7098333835601807, "target_z": -0.060546883952020604, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.6006026268005371, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/90a295a2d56fb83f7accf94b4a30b3cbabd1771c69c8e7572f7aab51c46e3984.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "9103087a07e74035d643b82b99691cd4f3dd410b82cf12e56d1d81d165510cdc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04288bb5a99a99037a9be28a427461d86d58b11967447c9666ff32f1aa389620", "generation_seconds_batch": 24.10760329803452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3148258924484253, "target_z": 0.41406246419191756, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.7206155061721802, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9103087a07e74035d643b82b99691cd4f3dd410b82cf12e56d1d81d165510cdc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "91f121864a162140661165e5267c6be32e9003b26bffdfffd48b52dcf6edeb7e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abf2361312e61839ce890354abfdcf3f2640778c1405820fa971556403d2b68a", "generation_seconds_batch": 26.04301957192365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9617929458618164, "target_z": -0.36328130371212364, "ce": 6.213237762451172, "genuineness": 4.645420074462891, "identity": 0.7644020318984985, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/91f121864a162140661165e5267c6be32e9003b26bffdfffd48b52dcf6edeb7e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "9256cc4da1d2799b5cd04412d7c5c06d38ca8a9f14da777645c2518bfbe0aa3d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be3470b2d2bb131749b6fd50603a41977203d7e18b0c3d893ed81e9485da11c7", "generation_seconds_batch": 17.83852531807497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9702102541923523, "target_z": 0.828125, "ce": 5.510311126708984, "genuineness": 4.419482707977295, "identity": 0.7559500336647034, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "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_TENS_reg_low/9256cc4da1d2799b5cd04412d7c5c06d38ca8a9f14da777645c2518bfbe0aa3d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "92af1bb8f9e3ff1c196d3058325db58e09dbe15a9b2817b03e9756bbef0b5cd1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d41de0dcd0e57d3bbe2dad0ef96f42dfb56214821d1cb1a38e112e95f6324498", "generation_seconds_batch": 24.186294093960896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4188607931137085, "target_z": 0.28906246419191756, "ce": 6.340320587158203, "genuineness": 4.532451152801514, "identity": 0.47114914655685425, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/92af1bb8f9e3ff1c196d3058325db58e09dbe15a9b2817b03e9756bbef0b5cd1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "92d39ccb599c6478238432a8f23ccd9452aceb970e913fae773852f253a5a17b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7f481af4a5770639c2cf5dce0f18a1f43acb9531ab5243fe07a59251b3669fa6", "generation_seconds_batch": 18.314719072077423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8081789016723633, "target_z": -0.17871097778409273, "ce": 6.141753673553467, "genuineness": 4.149612903594971, "identity": 0.5863714218139648, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/92d39ccb599c6478238432a8f23ccd9452aceb970e913fae773852f253a5a17b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "92eefba6b28316b74d2ec243e6a1d73113509af34704b5306ef62f4c1f32c51a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82c3a07d760a99e1cb2eaf285b9761e2ac8df54652cb84fa6072267eb08406f3", "generation_seconds_batch": 22.755767591064796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.978048324584961, "target_z": -0.38281246419191756, "ce": 6.133810997009277, "genuineness": 4.287685871124268, "identity": 0.6413828730583191, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/92eefba6b28316b74d2ec243e6a1d73113509af34704b5306ef62f4c1f32c51a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "930168f36ed3179892368167e9745499d66e631154de77e56126aeeb1f18d0e3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "134c3b00d16e87b2c6629c2f72f85c05c523e746c18f5be830a92bf7130a452b", "generation_seconds_batch": 17.27268993796315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.160109519958496, "target_z": -0.6015626074242473, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.735603928565979, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/930168f36ed3179892368167e9745499d66e631154de77e56126aeeb1f18d0e3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "934bd8096b6ce6fcc3234564cd7d4fe354e2cf8a6f378060d0011d10f64a9fa9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a5768bd10624dc2c07889049e0a2ea7dc06aaa44339d77ee223c6215c718d859", "generation_seconds_batch": 18.63979487202596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6293689012527466, "target_z": 0.03613279011994848, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.5383918881416321, "hypothesis": "Die Tür 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_TENS_reg_low/934bd8096b6ce6fcc3234564cd7d4fe354e2cf8a6f378060d0011d10f64a9fa9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "93736406ec6d2f3f3bca41a1a3a44b916c141f66f3e6ded3c7977df1ef5e813c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b73943e4b2160e41069da70882b79ce90038cd4e3c1abd6ede35158d377b2f2b", "generation_seconds_batch": 24.38807904999703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.017061471939087, "target_z": -0.42968753580808244, "ce": 6.260893821716309, "genuineness": 4.074300289154053, "identity": 0.17059338092803955, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/93736406ec6d2f3f3bca41a1a3a44b916c141f66f3e6ded3c7977df1ef5e813c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "93ea6564693f5bf8eca600e56c04bbf54aa830223321014dc1df1a640c47eaec", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a806d24b2ef8d2a5a5a7cea499b6d886d81ac58ae51ebb4bc99e920249be47b7", "generation_seconds_batch": 26.856478002970107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2969448566436768, "target_z": 0.4355468839520206, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.6875056028366089, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/93ea6564693f5bf8eca600e56c04bbf54aa830223321014dc1df1a640c47eaec.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "93fc337741a3e7bc9e06ae4b14a1c0e11364c4a2a7119117b90033050aa4a154", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4085b4342dfbcbed47b8893c3c4f402abbed88198d4118825dab079f2e20df24", "generation_seconds_batch": 22.770106348907575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4058563709259033, "target_z": 0.30468753580808244, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.34960097074508667, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/93fc337741a3e7bc9e06ae4b14a1c0e11364c4a2a7119117b90033050aa4a154.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "9420c86cfa7a5270f831e07779e68944da9f5cb00c7d887ad89fbe1e610d0020", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2600af23149423d13da4e2aacfa3b3b22f40b929dc2955b95fd06f69bf54336c", "generation_seconds_batch": 17.04239165200852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8325620889663696, "target_z": -0.20800786173611333, "ce": 5.95112943649292, "genuineness": 4.111956596374512, "identity": 0.6334262490272522, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9420c86cfa7a5270f831e07779e68944da9f5cb00c7d887ad89fbe1e610d0020.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "94379c598d6f61a1515e614f6f01956c87282648133085a6cb28a05f8f11b19b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f5aaadf90d080e3f616cae238482e04a846dab810b55107f8f923975e365b38", "generation_seconds_batch": 17.28567596094217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9552907943725586, "target_z": -0.3554688395202061, "ce": 6.221180438995361, "genuineness": 4.444586753845215, "identity": 0.5575391054153442, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/94379c598d6f61a1515e614f6f01956c87282648133085a6cb28a05f8f11b19b.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": "9450a4bf8199e4f32d6cfca31436eb80f051778276faa0a0083ad08860d19e97", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "9f6ed8199bf2522cabbff0c4063c094684a5d4b5790194888b055e12d6f07d7c", "generation_seconds_batch": 21.55353422590997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5944197177886963, "target_z": 0.07812492838383514, "ce": 6.324435234069824, "genuineness": 4.582659721374512, "identity": 0.2937089800834656, "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": 0.120000000000001, "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_TENS_reg_low/9450a4bf8199e4f32d6cfca31436eb80f051778276faa0a0083ad08860d19e97.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "94a3306cb78afa4a3f1eeb0279958b702d85441ab484e27e8e3295e827a3c2e3", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9a50db9c5c0a9aed5489b54e537817bf6bcfe5c49b506dfae56986b01b3e48d5", "generation_seconds_batch": 19.423177074058913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.917903184890747, "target_z": -0.3105468839520206, "ce": 6.1020402908325195, "genuineness": 4.093128204345703, "identity": 0.5008722543716431, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/94a3306cb78afa4a3f1eeb0279958b702d85441ab484e27e8e3295e827a3c2e3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "94be8a0802c20c5832f5873b2d305344e4c1611d2ec92c401ff4a7af74cba9bc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0d767d722d3392084cac3ec77fc55e7a9d4dd5b4fd2dfa8a6bdb2f5275e6ff4", "generation_seconds_batch": 25.65446751692798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6407476663589478, "target_z": 0.022460977784092732, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.6483989953994751, "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_TENS_reg_low/94be8a0802c20c5832f5873b2d305344e4c1611d2ec92c401ff4a7af74cba9bc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "94e25dcad0392ec80419f7f7f1082c7af3d212d15a65c5589262710f816e8666", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "7161ef6feb10a62a926bd2c6bb8002bea4f12f62eaf78ff16e8dc0555bb2aeea", "generation_seconds_batch": 20.382125031086616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.997554898262024, "target_z": -0.40625, "ce": 6.252951145172119, "genuineness": 4.494794845581055, "identity": 0.6707534790039062, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/94e25dcad0392ec80419f7f7f1082c7af3d212d15a65c5589262710f816e8666.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "94f5de2311b67dc5329014f0808aae3c4fe37e063c9890827e6f5e4d13b05ea2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7821342e320a95d2b112f0f48d7be914b284e22d339eea9a847d37418b302bf", "generation_seconds_batch": 22.193142421077937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7090206146240234, "target_z": -0.05957032592803091, "ce": 6.24500846862793, "genuineness": 4.042920112609863, "identity": 0.32003718614578247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/94f5de2311b67dc5329014f0808aae3c4fe37e063c9890827e6f5e4d13b05ea2.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": "95bac474f2de5c3a8ecc3a4409f099a7fbe4eb0f3c0831a354f980b0aa8a3149", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "c3af0b5193fc8c386512dd8a5e14766145e2420246aad6b3c6078495f457dbc6", "generation_seconds_batch": 21.017601352999918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3798476457595825, "target_z": 0.33593753580808244, "ce": 5.7684478759765625, "genuineness": 4.6956281661987305, "identity": 0.825592041015625, "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_TENS_reg_low/95bac474f2de5c3a8ecc3a4409f099a7fbe4eb0f3c0831a354f980b0aa8a3149.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": "95e5f32ef1a1b6371fce54e8aa68b40e219c2797b3a6b026ef1fb94071e2676e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "dbcee5e3f69a0103b35879e6cf340a0ddaea33d8507b64392e6a0aa293e57824", "generation_seconds_batch": 17.743225551908836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3570899963378906, "target_z": 0.36328130371212364, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.3841189742088318, "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_TENS_reg_low/95e5f32ef1a1b6371fce54e8aa68b40e219c2797b3a6b026ef1fb94071e2676e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "9605ea5a937be1f483c422d5e0cb1503808f0871d9896772af7a3d0fd1545c66", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfa7afc335f4653d5453fc23def9f05a6601258d4f21b52bc4659fc685735968", "generation_seconds_batch": 18.036964202998206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5757259130477905, "target_z": 0.10058590616792787, "ce": 5.808161735534668, "genuineness": 4.3127899169921875, "identity": 0.6241135597229004, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9605ea5a937be1f483c422d5e0cb1503808f0871d9896772af7a3d0fd1545c66.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": "961a3912f1412e7cc53dee5c4845427a5724a6786d96da83bda739ba6c92f5ff", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9d720fec16c99c1435cd947244ce4f2ce2e8d5bb59c09410601bb5c04676033a", "generation_seconds_batch": 21.190567411947995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4481205940246582, "target_z": 0.2539062320959588, "ce": 5.847874641418457, "genuineness": 4.68307638168335, "identity": 0.721836268901825, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/961a3912f1412e7cc53dee5c4845427a5724a6786d96da83bda739ba6c92f5ff.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": "963979ac3de7df4b8b63ae57c7877560516d27d03ed6a7e4ff75990cdf161658", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "992e8dc6f9d0176e79516dbb1afb51467e55327a2144a5fc25453da0d52c2064", "generation_seconds_batch": 24.477823984925635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3180769681930542, "target_z": 0.4101562320959588, "ce": 5.95112943649292, "genuineness": 4.444586753845215, "identity": 0.7896430492401123, "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": 0.120000000000001, "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_TENS_reg_low/963979ac3de7df4b8b63ae57c7877560516d27d03ed6a7e4ff75990cdf161658.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": "9663b14c49018cf02840d6f021bbca3805c32c052b1d9b9c709a1373d9ca592d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fd6ab4b22c53ffc58b9e5babbdfd1131027646b4ca00d618d92e904b3122ff8", "generation_seconds_batch": 24.882101555005647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6947970390319824, "target_z": -0.04248041727588151, "ce": 6.332377910614014, "genuineness": 4.6956281661987305, "identity": 0.6531136631965637, "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_TENS_reg_low/9663b14c49018cf02840d6f021bbca3805c32c052b1d9b9c709a1373d9ca592d.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": "966a3e6a6e0447f29512cc674f58732a3592c416dcf68947f1a3862ee38d9f44", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69b3dc434d4b82fbbec20bdaae179b01268c0d35d8e0f39bd69b9086ea2bd639", "generation_seconds_batch": 27.74315555789508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5489044189453125, "target_z": 0.13281246419191758, "ce": 6.340320587158203, "genuineness": 4.632867813110352, "identity": 0.23689685761928558, "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_TENS_reg_low/966a3e6a6e0447f29512cc674f58732a3592c416dcf68947f1a3862ee38d9f44.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "9684a2741ff2f0d8681b36da113730a40aab1060551e16f28f02dfbef75cba41", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ca4032c4b2d28141f27996e04fe941289542fbbd2f79654f49562e36b0b98c40", "generation_seconds_batch": 15.804846744984388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4332010746002197, "target_z": -0.9296875358080824, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.7225513458251953, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9684a2741ff2f0d8681b36da113730a40aab1060551e16f28f02dfbef75cba41.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "969091ea983694e6a71de42b4012d46798952c2bd9012efd9f02973fcf88f861", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "290515a9949e1e8f24ba7c4bfa15098ba11333258cf1cfddd2a3873deb610644", "generation_seconds_batch": 24.43980036000721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.693171501159668, "target_z": -0.040527301227902116, "ce": 6.340320587158203, "genuineness": 4.6203155517578125, "identity": 0.4527086317539215, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/969091ea983694e6a71de42b4012d46798952c2bd9012efd9f02973fcf88f861.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "96f70270095b2819147fd1902b63b0ed0fc2b49beab7d2f60325d0326a67f766", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b0499909bfe7949444365e5285a3e1701b7b3fa8a153db425311b21d445675b7", "generation_seconds_batch": 24.416452731005847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8008639812469482, "target_z": -0.16992195556818546, "ce": 5.919358730316162, "genuineness": 4.607763767242432, "identity": 0.7356692552566528, "hypothesis": "J'ai trouvé la da petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/96f70270095b2819147fd1902b63b0ed0fc2b49beab7d2f60325d0326a67f766.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": "9771d54f1aeb1403a3014b314bf9e4ede77fdc15ce352ab00016105ae27f3a85", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02bd29c68616bffae4f2aa0f882e9a5cab5a57d12ac00eb47f8d30d65594c170", "generation_seconds_batch": 19.658320095972158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.63424551486969, "target_z": 0.030273441976010302, "ce": 5.959072113037109, "genuineness": 4.369274139404297, "identity": 0.6655063629150391, "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_TENS_reg_low/9771d54f1aeb1403a3014b314bf9e4ede77fdc15ce352ab00016105ae27f3a85.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": "97a61f727da94c1eb02a5d5938a87caa7e705248d0d44b750d4d7ba89a457a08", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "b4d2978434dfbe6c0dcfa66ebafc341161d5ca10add82bd2e7bbaba4854c5c5b", "generation_seconds_batch": 18.859499237965792, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.825247049331665, "target_z": -0.19921869628787636, "ce": 5.855817794799805, "genuineness": 4.607763767242432, "identity": 0.6818047761917114, "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_TENS_reg_low/97a61f727da94c1eb02a5d5938a87caa7e705248d0d44b750d4d7ba89a457a08.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": "98828646dfab668cd8c9eef57a38cc04eb3148bcc5652c027c71025d5a399193", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e436565b2425a506e94e20975f83d87b2ff904ca8110dd9040b579e469b95929", "generation_seconds_batch": 26.67380578396842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7651019096374512, "target_z": -0.1269531160479794, "ce": 6.38797664642334, "genuineness": 4.432034492492676, "identity": 0.24475127458572388, "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_TENS_reg_low/98828646dfab668cd8c9eef57a38cc04eb3148bcc5652c027c71025d5a399193.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "98e15b3d46b773e112996123427f5ff1531f658d77257e41720abaf7b3db01a3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e2c0c427b8bce4750bd4605d0fa5e123da8b4e201c0a8f5a65cfb685d26da97", "generation_seconds_batch": 26.8028804929927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6218507289886475, "target_z": 0.04516602345801803, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.34529221057891846, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/98e15b3d46b773e112996123427f5ff1531f658d77257e41720abaf7b3db01a3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "98fb72c44c41112e75cc2be9a6058bd2f730a032398049259dc91214ef88efc3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "948db2e9da624da376aba4832eff22d2ae3cad110cc4662fe2441f874507fcda", "generation_seconds_batch": 24.33025464101229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2042887210845947, "target_z": 0.5468750716161649, "ce": 6.173524379730225, "genuineness": 4.2374773025512695, "identity": 0.853186845779419, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/98fb72c44c41112e75cc2be9a6058bd2f730a032398049259dc91214ef88efc3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "99489fb789457f79b7bc2e91e5760489d737fc616a4fa2b02d61a0731b6faae0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a6e46f196902cd948630e66524cd43791d117f25d7846fba834a5e7a974592d", "generation_seconds_batch": 25.97847821598407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2172931432724, "target_z": 0.53125, "ce": 6.117925643920898, "genuineness": 4.507347106933594, "identity": 0.7950279116630554, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/99489fb789457f79b7bc2e91e5760489d737fc616a4fa2b02d61a0731b6faae0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "996a9498d0de60cd6fc1564c05e18940f6b2839c61c03fd010516d2639946fc0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6fe26346a6ee9c574a6444dd9910a3af8e8299b5319467242e952e6577b8ae6", "generation_seconds_batch": 28.165058269980364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9455374479293823, "target_z": -0.34375, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.4937407374382019, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/996a9498d0de60cd6fc1564c05e18940f6b2839c61c03fd010516d2639946fc0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "99733b38868d929c95bae59339e93bc1963a1d4fc7cfc0c3bec7a8c1cb98e88b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a2808c557d51bc6bb36fb884fc6f48b8595c2b2c3a0e1279c789c4c7019c508f", "generation_seconds_batch": 20.307066318928264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8496302366256714, "target_z": -0.22851558023989696, "ce": 6.459460258483887, "genuineness": 4.432034492492676, "identity": 0.37766966223716736, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/99733b38868d929c95bae59339e93bc1963a1d4fc7cfc0c3bec7a8c1cb98e88b.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very relaxed and unstrained. 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": "7dc231c67a7eb80466d25aa39006aa3308794257212efa65f099efac7f4cca19", "generation_seconds_batch": 18.48532224004157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8740134239196777, "target_z": -0.25781246419191756, "ce": 6.372091293334961, "genuineness": 4.582659721374512, "identity": 0.5757800340652466, "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_TENS_reg_low/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "99d70dd6bde0cb09dc6fc630b3dbc9f5852225262285c0dd09ddf9fc1a7bbda1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e07cd03c9fa798648920afc2de6f713e78359e8931e01616b5a625addc758fa1", "generation_seconds_batch": 23.636896810028702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4383673667907715, "target_z": 0.26562492838383517, "ce": 6.054384231567383, "genuineness": 4.105680465698242, "identity": 0.49563175439834595, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/99d70dd6bde0cb09dc6fc630b3dbc9f5852225262285c0dd09ddf9fc1a7bbda1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "9a40bd15a48c164edfb8c3346ff61979a4b474bf79105f07259d07feab8f78aa", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "17dd305489af0072718b2afa46fc3b960662f8cadd953c25457579a2b6e6a2c9", "generation_seconds_batch": 16.692984931985848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4660015106201172, "target_z": 0.23242195556818546, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.6280828714370728, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9a40bd15a48c164edfb8c3346ff61979a4b474bf79105f07259d07feab8f78aa.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": "9a7bf59d3966d468b2358ef0de9844eeeb06bf4565df75960eca4221bdb00da0", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8951890ecd37f97a7b38d09a37d2d980f446e94b1cf3fbf72730523ebbabe227", "generation_seconds_batch": 20.038881281972863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8967710733413696, "target_z": -0.2851562320959588, "ce": 6.475345611572266, "genuineness": 4.262581825256348, "identity": 0.37057358026504517, "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_TENS_reg_low/9a7bf59d3966d468b2358ef0de9844eeeb06bf4565df75960eca4221bdb00da0.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": "9afb4916c7c36230c1034fb066dad39be08e36e927a421e21ebb8eaecc59fa31", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "612c913594fd1df3b8773c399ddcb884e9e2dfe0ad3bca76fe759911d52e0449", "generation_seconds_batch": 20.279402674990706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.52045738697052, "target_z": 0.16699213826388667, "ce": 5.927301406860352, "genuineness": 4.68307638168335, "identity": 0.6558467149734497, "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_TENS_reg_low/9afb4916c7c36230c1034fb066dad39be08e36e927a421e21ebb8eaecc59fa31.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "9b297d44480ed607758a63aed433e68a3eb627958e778512a29c7bf8fd94a2ac", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec6dfbcf2f02b75db128d67621fe0bd20e1084ce574a3ee8336aed756e3999d4", "generation_seconds_batch": 23.243780029937625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6626925468444824, "target_z": -0.0039062320959587856, "ce": 5.863759994506836, "genuineness": 4.896461486816406, "identity": 0.7936007976531982, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9b297d44480ed607758a63aed433e68a3eb627958e778512a29c7bf8fd94a2ac.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "9b6fdcd208114d95b30a7e82811a0aab11fd715dd319c9b6df0b9e7f4bd7cb47", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7dbef554d30df3d60eccad1f7eaf75a9664b17abb2cf9ac0f1b36171182e9e1", "generation_seconds_batch": 24.677046103053726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0365679264068604, "target_z": -0.45312492838383517, "ce": 6.41180419921875, "genuineness": 4.65797233581543, "identity": 0.44666731357574463, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9b6fdcd208114d95b30a7e82811a0aab11fd715dd319c9b6df0b9e7f4bd7cb47.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very relaxed and unstrained. 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": "6793adfc23f20ec5f7a6a4ee002d9ecc12ad8d4865f7e4854355bb9eae45af3f", "generation_seconds_batch": 17.750948000932112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9130265712738037, "target_z": -0.30468753580808244, "ce": 6.324435234069824, "genuineness": 4.494794845581055, "identity": 0.7885874509811401, "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_TENS_reg_low/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "9c3d252470c8089253123892355753091fa92b33ab8ed27bc6d1fe2ed3bc3ef4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "6122b59784753e5c2ec817d32c4bdea847d710218f73321d4a18ada654ed7333", "generation_seconds_batch": 26.658779768040404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.04823637008667, "target_z": 0.7343750716161649, "ce": 5.935244083404541, "genuineness": 4.607763767242432, "identity": 0.7875509262084961, "hypothesis": "Ah j'ai trouvé la petite boite. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble.", "wer": 0.2631578947368421, "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_TENS_reg_low/9c3d252470c8089253123892355753091fa92b33ab8ed27bc6d1fe2ed3bc3ef4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "9c68228e4e85d23a6ecabc94d3e7b2141dc22e70b57466517602ae841c92bb7a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9594b6aee1e3b11bd222cfe60b381c11a3cf2f563067a7eafba6c5f1a83a8a1f", "generation_seconds_batch": 25.782550566014834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0755810737609863, "target_z": -0.5, "ce": 6.1894097328186035, "genuineness": 4.425758361816406, "identity": 0.5448873043060303, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9c68228e4e85d23a6ecabc94d3e7b2141dc22e70b57466517602ae841c92bb7a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "9c78b8eb994c4eb5c065557909974e9137255f72314d90877ce6fec7a3145e21", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7be30050d434e60f168f6757eb37520df44259b41c17da5a1a92d6dac3797885", "generation_seconds_batch": 16.754659287980758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.76835298538208, "target_z": -0.13085934814393818, "ce": 6.268836498260498, "genuineness": 4.086852550506592, "identity": 0.375005304813385, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9c78b8eb994c4eb5c065557909974e9137255f72314d90877ce6fec7a3145e21.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "9c8adc83a6773e0e2efc92b65d5712bcbc4d272c74ffa352c24a2d188cbed443", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "886144f88c344b50b98e715a824d8bf21f328105efd19ca62c71cf2bdc2f16a7", "generation_seconds_batch": 22.75438572803978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6795575618743896, "target_z": -0.024169882709909843, "ce": 6.443574905395508, "genuineness": 4.444586753845215, "identity": 0.7050633430480957, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9c8adc83a6773e0e2efc92b65d5712bcbc4d272c74ffa352c24a2d188cbed443.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very relaxed and unstrained. 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": "a8843faeba57ccc15d25d89b4efd9ba0c004816e45d4d05cff0df2d4057f6f3f", "generation_seconds_batch": 17.718867521965876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.031691312789917, "target_z": -0.447265580239897, "ce": 6.09409761428833, "genuineness": 4.570107460021973, "identity": 0.6340510845184326, "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_TENS_reg_low/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "9cdb3d339d21982267086d54b00cb63e09e94bae7265b8576b1a7bfdc5695a84", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a10e442751a28704ecc646515f490bdeb70cbfc8afa9c5dffdd1942504c1a273", "generation_seconds_batch": 19.724639949039556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6832150220870972, "target_z": -0.02856439381786348, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.6482461094856262, "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_TENS_reg_low/9cdb3d339d21982267086d54b00cb63e09e94bae7265b8576b1a7bfdc5695a84.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "9d022faf6c3bb2f29cb2daf3ba8c626e1af84b87362ea5272936b00dc970db75", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56f64dbed2016cefb84024db9db95a33e4b7af9bb12f80dad48625e0058f5a71", "generation_seconds_batch": 27.577563275001012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.848004698753357, "target_z": -0.22656246419191758, "ce": 6.062326908111572, "genuineness": 4.375550270080566, "identity": 0.37245696783065796, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9d022faf6c3bb2f29cb2daf3ba8c626e1af84b87362ea5272936b00dc970db75.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "9d85dc8b16e2e63a4b81ec0561d8ef5f888b9b53da219f9ccd399e3f123efb60", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52f4bf00d52fd3a60cfce97986037c860eb91e638da49f1c505da199db9bd2c7", "generation_seconds_batch": 19.69857766502537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5862919092178345, "target_z": 0.08789065185606182, "ce": 5.990842819213867, "genuineness": 4.545003414154053, "identity": 0.6474943161010742, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9d85dc8b16e2e63a4b81ec0561d8ef5f888b9b53da219f9ccd399e3f123efb60.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": "9e06e890f2491d8ad7ab7316e6f67e8368c313528e75f9f65ce5cac97759d0c4", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa26d7663db662242bf97c102d811db3a5a60110da167cbcbbe939c2d0888b55", "generation_seconds_batch": 23.830914153018966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4464950561523438, "target_z": 0.2558593481439382, "ce": 5.784333229064941, "genuineness": 4.557555198669434, "identity": 0.8273981809616089, "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_TENS_reg_low/9e06e890f2491d8ad7ab7316e6f67e8368c313528e75f9f65ce5cac97759d0c4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "9ee49ff5c6f8f003095c78ad16a8734a0df8fbca2af76dff8cf64a0ef5233ef8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "919363fadb6d173658b5dea2122c6a3aaf8504b4f39fca7481226f075a732300", "generation_seconds_batch": 17.962101444019936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.702112078666687, "target_z": -0.05126958272411849, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.6256246566772461, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9ee49ff5c6f8f003095c78ad16a8734a0df8fbca2af76dff8cf64a0ef5233ef8.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": "9ee8d5f0fc0ed77c15aca524bc0a609321fe351f1d8d5717fc825dae96c1788a", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "154f23569e4874b9550970626ff907eaf2a7331519ba307ae8dcf2eaf4c25515", "generation_seconds_batch": 17.95438067591749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6643180847167969, "target_z": -0.0058593481439381786, "ce": 6.41180419921875, "genuineness": 4.494794845581055, "identity": 0.4052217900753021, "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_TENS_reg_low/9ee8d5f0fc0ed77c15aca524bc0a609321fe351f1d8d5717fc825dae96c1788a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "9ef0547998fcc7e9dffea614af1ce2937a7890123b15981db24e6c61b90d3b82", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e17903910d4c1c11be9786de4f5931ce6ffbc9295858b2be48a2d0f5255b1ef2", "generation_seconds_batch": 22.85814471996855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8154938220977783, "target_z": -0.1875, "ce": 6.292664527893066, "genuineness": 4.356721878051758, "identity": 0.20284715294837952, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9ef0547998fcc7e9dffea614af1ce2937a7890123b15981db24e6c61b90d3b82.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "9f09f5ddc78e4b448ecf53be1521db7fee150bf316870678417e526f821a91fa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "993405c7e07be9e37e4ba6770f0e99dc1d6e729988fd7f8c0d6db1f5a61fd4a5", "generation_seconds_batch": 19.03227954893373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6066112518310547, "target_z": 0.0634765580239897, "ce": 6.38003396987915, "genuineness": 4.457139015197754, "identity": 0.3930807411670685, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9f09f5ddc78e4b448ecf53be1521db7fee150bf316870678417e526f821a91fa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "9f0a2023149302a4bb5a8709a01ef6b29d7d568c0ad59cf77eba88f97bbffdbd", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78bc6a54d605607ed9341cf7ff6d9ffa3a728b71165a8e824eddfd9122155991", "generation_seconds_batch": 23.9797532169614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1893692016601562, "target_z": -0.6367186962878764, "ce": 5.911416053771973, "genuineness": 4.375550270080566, "identity": 0.7628068327903748, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9f0a2023149302a4bb5a8709a01ef6b29d7d568c0ad59cf77eba88f97bbffdbd.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": "9f844d3c5ddf825bf3290e682873d99711ce78f54d14ef49b4f5ebd76b580dc2", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bd6de35890f7626e524087c96e9bfeeef919a034f55cb34f398544a5d8782d4", "generation_seconds_batch": 24.237278033979237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2270464897155762, "target_z": 0.519531160479794, "ce": 5.681078910827637, "genuineness": 4.645420074462891, "identity": 0.7458933591842651, "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_TENS_reg_low/9f844d3c5ddf825bf3290e682873d99711ce78f54d14ef49b4f5ebd76b580dc2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "9fa5a1c0578b3ae6af5b780c196c9bb742e608d34740b11914e492f854a38e48", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "918a6991be1dc2191678db39e013d382bbfef12373df4efc96a8f376019ab1e2", "generation_seconds_batch": 22.601890597026795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1048407554626465, "target_z": -0.5351560888636291, "ce": 6.1020402908325195, "genuineness": 4.306513786315918, "identity": 0.5788944959640503, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/9fa5a1c0578b3ae6af5b780c196c9bb742e608d34740b11914e492f854a38e48.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "a02bcbef3d73232514ba34ca7acea9a3bcb97cc63d49647b95f7f95c6f16a9dc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c9d2fc53bd07b42db095673cdfd2a5b76d316676cafbaa6fb9fefae195adb7f", "generation_seconds_batch": 19.65862367209047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4156097173690796, "target_z": 0.29296869628787636, "ce": 6.332377910614014, "genuineness": 4.400654315948486, "identity": 0.4619380235671997, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a02bcbef3d73232514ba34ca7acea9a3bcb97cc63d49647b95f7f95c6f16a9dc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "a0476daf52da959fc2fb6d272e156da6e07cfa21efcc4fce87039a8bac04de6d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a92f1aa7d072e62cf8bb254486b86f3c8824058dcba9eb413816df19a34867f", "generation_seconds_batch": 23.229884275933728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.308323621749878, "target_z": 0.42187507161616483, "ce": 5.518253803253174, "genuineness": 4.645420074462891, "identity": 0.530251145362854, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a0476daf52da959fc2fb6d272e156da6e07cfa21efcc4fce87039a8bac04de6d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "a086d727a0a58438356553cda8d8906f6d029518ad75469789c59ef5fb4cc035", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "908fa73b7cf7b40bf0229a8f69ba6af23a58e8dd9d247755fa94df8beb4566f1", "generation_seconds_batch": 20.05072914902121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5822280645370483, "target_z": 0.0927734419760103, "ce": 6.054384231567383, "genuineness": 4.300237655639648, "identity": 0.6855831742286682, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a086d727a0a58438356553cda8d8906f6d029518ad75469789c59ef5fb4cc035.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "a0c9788671e4b2f27588c996e5302d7cc241c3324571df98d7e58b057325b54f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "92dfdfd3ab9081631fc6602b5a517b28040629f1365536a6c783630746463412", "generation_seconds_batch": 18.090922074974515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2985703945159912, "target_z": 0.4335937679040412, "ce": 6.205295085906982, "genuineness": 4.6203155517578125, "identity": 0.8429157733917236, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a0c9788671e4b2f27588c996e5302d7cc241c3324571df98d7e58b057325b54f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "a0f6bafd7512af95bda777e217bd0539219326c73bd8a6abb3aeb7cd24454c45", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbba2407bd260680409a0da79444b1c3b451f37a79cf4020b0b65d7a0783e34e", "generation_seconds_batch": 23.620418836013414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.787859559059143, "target_z": -0.1542968839520206, "ce": 6.292664527893066, "genuineness": 4.369274139404297, "identity": 0.49355724453926086, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a0f6bafd7512af95bda777e217bd0539219326c73bd8a6abb3aeb7cd24454c45.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "a152014cfa1a3d1ed951e55d57b25c58ed839ce45a1e5db276d3dbfdc2699d90", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33ce97577eb46a277011cfd9bc2e5fff19bb3d3630dd3a2cd3bcbdcf437ad71f", "generation_seconds_batch": 18.63979487202596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.508265733718872, "target_z": 0.18164065185606182, "ce": 5.903473377227783, "genuineness": 4.406930446624756, "identity": 0.5564566254615784, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a152014cfa1a3d1ed951e55d57b25c58ed839ce45a1e5db276d3dbfdc2699d90.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": "a1523addb05e95c8689b4ae5deae87f6001d1fe96c64c51285216fb90cbf21af", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af560ec4822e449370749f3e03e6f56a31f5736f6f039f6165234b9f4e5f26dd", "generation_seconds_batch": 26.38442269200459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3066980838775635, "target_z": 0.42382818766414426, "ce": 6.038498878479004, "genuineness": 4.507347106933594, "identity": 0.585159420967102, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a1523addb05e95c8689b4ae5deae87f6001d1fe96c64c51285216fb90cbf21af.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "a25362991ba20d641aab21954548ffa8914e6c616d24fd12ba922d7466bf431d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a84e41e9c7af2b5e57be5bbda04379681124f36808ac44c4d8356c39edf0c3af", "generation_seconds_batch": 16.976331437937915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.530733823776245, "target_z": -1.0468750716161648, "ce": 6.078212261199951, "genuineness": 4.331617832183838, "identity": 0.6300573348999023, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a25362991ba20d641aab21954548ffa8914e6c616d24fd12ba922d7466bf431d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "a2546845d59f7df782796c8c8b0f3764e7e11f595eea386fe16a4ccfd71f7457", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abd219d95c1a8cd03c06ca30b417ef81374cc6f15168ca9f16f31005c1e38916", "generation_seconds_batch": 23.917896409053355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5957555770874023, "target_z": -1.125, "ce": 6.149696350097656, "genuineness": 4.319066047668457, "identity": 0.7423180341720581, "hypothesis": "De deur staat nog open. Ik heb je ik heb je jas, ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a2546845d59f7df782796c8c8b0f3764e7e11f595eea386fe16a4ccfd71f7457.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "a31f60f78c6e1a9d7d0a51d86b63af1dfa0d890807a26d12aeabe9234865dba3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "bdfd5f2f0eadeed65f21d119c8579bc96d652ebe75e4491db35c172a0ffa8566", "generation_seconds_batch": 24.17586968606338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4863208532333374, "target_z": 0.20800786173611333, "ce": 6.316492557525635, "genuineness": 4.65797233581543, "identity": 0.7884296178817749, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/a31f60f78c6e1a9d7d0a51d86b63af1dfa0d890807a26d12aeabe9234865dba3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "a34b366214fc50bcc6ce001cfaa0657862bf1de32959c68394b65e859dfabb97", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0fd260bcb9e213f9610a5e3133f4f0d2910ebd2aa1ef8ba7ea0d2913fb0a4206", "generation_seconds_batch": 24.201527010067366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2010376453399658, "target_z": 0.5507813037121236, "ce": 5.879645347595215, "genuineness": 4.281409740447998, "identity": 0.3161332607269287, "hypothesis": "I found the little box. It was beside the old window. We can look inside", "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_TENS_reg_low/a34b366214fc50bcc6ce001cfaa0657862bf1de32959c68394b65e859dfabb97.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "a3c4a2701f2bffb0b9f850d3faac16d84fc4a6c09a0f82a840dd790be226da10", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "411a89f1146b77cb73e4bd3cf3f9a9d8a1b1afc73910a97d5171b16622bea899", "generation_seconds_batch": 23.706586930900812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4725037813186646, "target_z": 0.22460934814393818, "ce": 5.998785495758057, "genuineness": 4.193545341491699, "identity": 0.6513148546218872, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a3c4a2701f2bffb0b9f850d3faac16d84fc4a6c09a0f82a840dd790be226da10.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "a3e096982ec2c240d809114fca36b00cd684db21ef0abbac4e648f94b69b139f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf9bf26ed37dc55dfdec6cbd66313c2d3f4ec3629eb1a222bcc79efff025ded8", "generation_seconds_batch": 23.43494895193726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2075397968292236, "target_z": 0.542968839520206, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.8087993860244751, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a3e096982ec2c240d809114fca36b00cd684db21ef0abbac4e648f94b69b139f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "a42935805827cc915f43c5043e57521c23e6584ff930ff957871890a91665f48", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a39401053aa68cc0a8ea23aa42c7fa62bb10713df6d2dd55618c3e024e3b7dd", "generation_seconds_batch": 25.33220457204152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9894272089004517, "target_z": -0.396484419760103, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.7216765880584717, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a42935805827cc915f43c5043e57521c23e6584ff930ff957871890a91665f48.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": "a4c32cb3213f5176af23493493342884c4d62d9012ecbd4da81e831f4a7a570c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76cc52d547e64cf41cd8c07ac83e0f23b04cfec29fe3394e2d218dd891ea3f4e", "generation_seconds_batch": 17.181772356969304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6350582838058472, "target_z": 0.029296883952020607, "ce": 6.316492557525635, "genuineness": 4.7834930419921875, "identity": 0.6583519577980042, "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_TENS_reg_low/a4c32cb3213f5176af23493493342884c4d62d9012ecbd4da81e831f4a7a570c.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": "a4e46d51946fa129b2d794320bfcd59e0c6fa7f719d9e59ebaa39f683077b2eb", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "610c57bbb7877c72b504f87544f0eee15268dcafd32a8e8d7326e39df15a255d", "generation_seconds_batch": 19.92553039395716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.940660834312439, "target_z": -0.3378906518560618, "ce": 6.141753673553467, "genuineness": 4.595211505889893, "identity": 0.8167703151702881, "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_TENS_reg_low/a4e46d51946fa129b2d794320bfcd59e0c6fa7f719d9e59ebaa39f683077b2eb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "a4fb677d05b3b998fe0a456245cff7ded73261d0bbccad3e62920fc36dfebff5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3622d80c3a6ad4b265f261876ede877232a2366ee4e69239b95276bcf833556", "generation_seconds_batch": 19.288109980989248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5017635822296143, "target_z": 0.1894531160479794, "ce": 6.022613525390625, "genuineness": 4.231201171875, "identity": 0.6005727648735046, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a4fb677d05b3b998fe0a456245cff7ded73261d0bbccad3e62920fc36dfebff5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "a6665cabc4a7c8dd6212f23d2cd6fb18d8a93900e1c6f24d8dfefe6839f1bc93", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fddd3e89e07926d431ced090cb5ab66e8ec4c25189919a3262a70d4e1aa5590e", "generation_seconds_batch": 23.27146893402096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.730965495109558, "target_z": -0.08593753580808243, "ce": 6.356205940246582, "genuineness": 4.645420074462891, "identity": 0.4673716127872467, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a6665cabc4a7c8dd6212f23d2cd6fb18d8a93900e1c6f24d8dfefe6839f1bc93.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "a715a8461cb6236ecf471992099b776c3585ad0abdb9024c3fc3975695c50e9f", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "eb512d4ce59c40d4f10ee2d238150c3f8ec4ce72b96de9e68c93c7689045dac5", "generation_seconds_batch": 18.182881890097633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5399638414382935, "target_z": 0.14355474568813395, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.5192346572875977, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a715a8461cb6236ecf471992099b776c3585ad0abdb9024c3fc3975695c50e9f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "a74fdd5e7b2a89cd7a6456cb27b11fefd1e6668ecb867cca04cc76f18b4be86e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6089a51958b66308a6b66d7df169cd00f4ad9ae86c57c62caa7ba4cde22b5a21", "generation_seconds_batch": 18.07121021405328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7484400272369385, "target_z": -0.10693353332386091, "ce": 5.8875885009765625, "genuineness": 4.243753433227539, "identity": 0.6321900486946106, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a74fdd5e7b2a89cd7a6456cb27b11fefd1e6668ecb867cca04cc76f18b4be86e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "a7a6aa20d1d77ec8305ec8a154a43d1960feddc839ec0512a559863b3c80d3d4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "75a0ef58ddf1aa0c715b1610a71e031b2969f0af7bb2e2b437a68d5c92b10935", "generation_seconds_batch": 26.629237808985636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.763069987297058, "target_z": -0.12451172098800516, "ce": 6.046441555023193, "genuineness": 4.570107460021973, "identity": 0.5114495754241943, "hypothesis": "Eet, ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a7a6aa20d1d77ec8305ec8a154a43d1960feddc839ec0512a559863b3c80d3d4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "a7b0c611d9164334569b73b4dcc00eab1b7a4e89602045f3163b8e7ebe4d853d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a9f3a34c4193980a05088b77dabeb9ca444ea79423d4b90c4979f5186adab87", "generation_seconds_batch": 24.64905165194068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4603121280670166, "target_z": 0.23925786173611333, "ce": 6.054384231567383, "genuineness": 4.507347106933594, "identity": 0.7899646162986755, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a7b0c611d9164334569b73b4dcc00eab1b7a4e89602045f3163b8e7ebe4d853d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "a7ec596b1ea2f83b6377fdee6cefac95632693f6f20b5b4d7d8380e045af0095", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dde25a7ff6100054a5fce5133e57d8ee45332a4ee1eae7fd3a01222de0c1e7c", "generation_seconds_batch": 24.093465792015195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6090495586395264, "target_z": 0.060546883952020604, "ce": 5.95112943649292, "genuineness": 3.9550557136535645, "identity": 0.6997449994087219, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a7ec596b1ea2f83b6377fdee6cefac95632693f6f20b5b4d7d8380e045af0095.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "a81031a20d13c93f661f805b242fd3020596d19e5a321b70cdd3a3cdc32faff1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c2b3be50f79737bf5570a38ba5c26afef873de71a41b79f0c773ba0c1794447b", "generation_seconds_batch": 23.716581071028486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7529103755950928, "target_z": -0.11230474568813395, "ce": 6.38797664642334, "genuineness": 4.293961524963379, "identity": 0.5410382747650146, "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_TENS_reg_low/a81031a20d13c93f661f805b242fd3020596d19e5a321b70cdd3a3cdc32faff1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "a854e2ebfeb383e463fb569a55f84980f412171a7defddbd017b6f068cc07fec", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa53fd141e192b37c2e3100de7b0d6cc802c70f5e4675f7c36a78393e5b675d1", "generation_seconds_batch": 28.05661284807138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.044985294342041, "target_z": 0.7382813037121236, "ce": 5.764476776123047, "genuineness": 4.444586753845215, "identity": 0.8957512378692627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a854e2ebfeb383e463fb569a55f84980f412171a7defddbd017b6f068cc07fec.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": "a8a4b02d80feb877aca63c497c4f2b0c5b27f029f3596e5dc35b83618c224435", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3defd86c7a1ab9b8e22a748e4fbfce3374f512a73c1477427d21292d66e175c7", "generation_seconds_batch": 22.808008905034512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2611829042434692, "target_z": 0.478515580239897, "ce": 5.959072113037109, "genuineness": 4.7081804275512695, "identity": 0.7178864479064941, "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_TENS_reg_low/a8a4b02d80feb877aca63c497c4f2b0c5b27f029f3596e5dc35b83618c224435.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very relaxed and unstrained. 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": "85b944f4badb5a6c6b8f3da1e0ec3fc73986f1bd788c302cf42f67bba48e4e56", "generation_seconds_batch": 15.792407928034663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.956916332244873, "target_z": -0.35742195556818546, "ce": 6.125868320465088, "genuineness": 4.143336772918701, "identity": 0.7169149518013, "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_TENS_reg_low/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "a90199292ae3bdc7c28e03cfea5b7b9ea43c8807dc69bc8fbb41976bbcd41907", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "960d4ce31d14558638c87370af8187aae16a7b322def9d9ccd54b1c801d593fe", "generation_seconds_batch": 25.467353142099455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3603410720825195, "target_z": 0.35937507161616483, "ce": 6.1020402908325195, "genuineness": 4.770940780639648, "identity": 0.6767998933792114, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a90199292ae3bdc7c28e03cfea5b7b9ea43c8807dc69bc8fbb41976bbcd41907.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "a90ebd1c834ebf5e3cc3b2c59ac75b553ed4d94a6656168d3ba67c0e0d0acd65", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "035bdfc8b77e15314e551257994f3e81ca66c2d81faff299b1a8334da1dc08a6", "generation_seconds_batch": 23.706586930900812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3798476457595825, "target_z": 0.33593753580808244, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.8380489349365234, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a90ebd1c834ebf5e3cc3b2c59ac75b553ed4d94a6656168d3ba67c0e0d0acd65.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very relaxed and unstrained. 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": "52bc8cd5e0816c0a2b78305723c341af89bfb9969a02dfdae5ee2cfd6e3fd895", "generation_seconds_batch": 17.960630608955398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8740134239196777, "target_z": -0.25781246419191756, "ce": 6.284721851348877, "genuineness": 4.645420074462891, "identity": 0.7723166942596436, "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_TENS_reg_low/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "a99829370f78198aa5476be792603eb7aa976724e2e9e8eef89a18dfa1e79970", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d837b9865cd517c29bcb95084a4dac6faf43bbf6f8d0513b98396b3019e89ab5", "generation_seconds_batch": 22.347357863909565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7195866107940674, "target_z": -0.07226558023989696, "ce": 6.332377910614014, "genuineness": 4.306513786315918, "identity": 0.7559542655944824, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a99829370f78198aa5476be792603eb7aa976724e2e9e8eef89a18dfa1e79970.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "a9ccc3e2a6b82324df176cbe4d1fb10df8dde1a680fc626c147c921fe827fc32", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9bdc4e6e17e736e8d061eebcdb00c48181f01bbd84f9dbade064a30a56a10bb2", "generation_seconds_batch": 18.931981042027473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6872789859771729, "target_z": -0.03344732717014167, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.6509085297584534, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/a9ccc3e2a6b82324df176cbe4d1fb10df8dde1a680fc626c147c921fe827fc32.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "aa009dbd75d12e5a04848119734892c7fe633f236ccf49b56b8ef56c5f24d77c", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "02e039578adcee635b2b337a9465de04c9da0380fb505ba247b9f18c9ce1cf96", "generation_seconds_batch": 16.803423152072355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6968289613723755, "target_z": -0.04492181233585575, "ce": 6.197352409362793, "genuineness": 4.7081804275512695, "identity": 0.6263607740402222, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/aa009dbd75d12e5a04848119734892c7fe633f236ccf49b56b8ef56c5f24d77c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "aac238174375c94787c8e1fac271e2f5e40fc545f5d63e16e4903144d58802a2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "698c405033aee2ed5e897c3aa1687bf8812d6d4e0d4aadb0c7da87e861fedb4e", "generation_seconds_batch": 15.882174622034654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.974797248840332, "target_z": -0.3789062320959588, "ce": 6.221180438995361, "genuineness": 4.300237655639648, "identity": 0.6385456919670105, "hypothesis": "Ik heb het klein doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/aac238174375c94787c8e1fac271e2f5e40fc545f5d63e16e4903144d58802a2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "aaf05461543c787e3b02c04d862f6353e48609b254db014ce9a3a0915ef6192a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5b5a9b79a15c6bd53d02e09db0bbf04619b5fd4b6d6117ab2a48091ec456289b", "generation_seconds_batch": 18.502930148970336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6005154848098755, "target_z": 0.07080074320391241, "ce": 6.300607204437256, "genuineness": 4.758388519287109, "identity": 0.8493993878364563, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/aaf05461543c787e3b02c04d862f6353e48609b254db014ce9a3a0915ef6192a.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very relaxed and unstrained. 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": "e06766f709f8a367f407198d6097e0ab6bf8b99963ba6e819bd3806673068f77", "generation_seconds_batch": 17.68461528304033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.059325695037842, "target_z": -0.4804688395202061, "ce": 6.435632705688477, "genuineness": 4.168440818786621, "identity": 0.45995593070983887, "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_TENS_reg_low/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "ac036766b2da242bc5602e6e58014602fa7e6f4c0d448f7c7d8917300b178100", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd5d50a8bb3eccd899e357d3f319f73a08b6ccb760ee030d0959365495e1874e", "generation_seconds_batch": 26.04301957192365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6001091003417969, "target_z": 0.07128902221590727, "ce": 6.205295085906982, "genuineness": 4.18726921081543, "identity": 0.5364415645599365, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen.", "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_TENS_reg_low/ac036766b2da242bc5602e6e58014602fa7e6f4c0d448f7c7d8917300b178100.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "ac3090a086efa2fc60bb2b13885cd6337d84eaa1a71c7ae1f890b36e2068873e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdd015ebe695e2c536417bfaca3aaac0b7a1560f7cba2a8ae705f19bbca5d37a", "generation_seconds_batch": 16.976331437937915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7778167724609375, "target_z": -1.3437501432323298, "ce": 5.982900142669678, "genuineness": 4.457139015197754, "identity": 0.6059210300445557, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ac3090a086efa2fc60bb2b13885cd6337d84eaa1a71c7ae1f890b36e2068873e.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": "ac3b056d963a75d2229a7f7f76ed47b564199721ac7e39710ff4eb358ecd9341", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "84994d0072095920a1620e5aada3465d62c2b1d6982d555e5b19978300da71f6", "generation_seconds_batch": 27.74185279000085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6155517101287842, "target_z": 0.052734419760103034, "ce": 5.919358730316162, "genuineness": 4.582659721374512, "identity": 0.7156401872634888, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ac3b056d963a75d2229a7f7f76ed47b564199721ac7e39710ff4eb358ecd9341.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "aca8ac5891837345ae3c8e6eff58ccb5573b977eac00a5619ff71139a8593772", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bba94bcf837a2e897c20b1f6bfe1f066b57ed007a95e5336623209f158736278", "generation_seconds_batch": 23.332554060965776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7886723279953003, "target_z": -0.1552734419760103, "ce": 5.95112943649292, "genuineness": 4.971774101257324, "identity": 0.7120989561080933, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/aca8ac5891837345ae3c8e6eff58ccb5573b977eac00a5619ff71139a8593772.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "acdd4871d92a66fd5fd28d0b369327a06d24c15524fe21dd6a080daf8c505db2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1473889d5a07c5ecdc1ee9379ce8db8a74db96c6842151e5cc39aa325f515049", "generation_seconds_batch": 17.112515517976135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6157549619674683, "target_z": 0.05249020863794075, "ce": 6.356205940246582, "genuineness": 4.356721878051758, "identity": 0.3713105320930481, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/acdd4871d92a66fd5fd28d0b369327a06d24c15524fe21dd6a080daf8c505db2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "ad27879f6bea71f6e7a31605a5bcecb55714d3468216bda7f6dad774b1ff4db8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89a66a4c9169953f472696a01235ff3b3a2ea308298125cb1da68df3a7d85a74", "generation_seconds_batch": 18.369528984068893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4123585224151611, "target_z": 0.29687507161616483, "ce": 6.070269584655762, "genuineness": 4.381826400756836, "identity": 0.4548611342906952, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ad27879f6bea71f6e7a31605a5bcecb55714d3468216bda7f6dad774b1ff4db8.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very relaxed and unstrained. 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": "86244a591db2e99ddca1fc6be1cd1d77938eb13890c20b773ec946c66469da02", "generation_seconds_batch": 17.282651406014338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9552907943725586, "target_z": -0.3554688395202061, "ce": 6.046441555023193, "genuineness": 4.425758361816406, "identity": 0.5283641815185547, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las ze twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "add2143be61acb06da3c68bb894366af07fd8d054389888b9f16787f3c86c078", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d69892ad4735da5263ec621170d46320d9496b0538a0b7b5d164679db5654783", "generation_seconds_batch": 17.513587839901447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8886433839797974, "target_z": -0.2753906518560618, "ce": 6.356205940246582, "genuineness": 4.331617832183838, "identity": 0.23382525146007538, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/add2143be61acb06da3c68bb894366af07fd8d054389888b9f16787f3c86c078.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "ae216037d178b77d08d817eb336ed762fb12ca540d11b1b647417f312b53e618", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2551b0487935752f33f062c3605dcd8188b4cbf8bbdabf5f3decf46f06a8f43c", "generation_seconds_batch": 17.211615551961586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9845504760742188, "target_z": -0.39062492838383517, "ce": 6.268836498260498, "genuineness": 4.425758361816406, "identity": 0.5487175583839417, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ae216037d178b77d08d817eb336ed762fb12ca540d11b1b647417f312b53e618.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very relaxed and unstrained. 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": "f4bfe698acfa1aa361f4572120c13758011f1dd8a409b7d5b72368b194b3e3c0", "generation_seconds_batch": 21.618182509904727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8805155754089355, "target_z": -0.26562492838383517, "ce": 6.499174118041992, "genuineness": 4.300237655639648, "identity": 0.13268405199050903, "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_TENS_reg_low/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "ae6b389dcdfb6cce0b41b929271ef7d59e7ac456a2116122b73596159c4f0fb9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dd1cbe9747f82c70c48f5c1dd3bbd4f94680e230cdc047a37c11ec26216b72d8", "generation_seconds_batch": 24.897381274960935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9715461730957031, "target_z": -0.375, "ce": 4.505563259124756, "genuineness": 4.218649387359619, "identity": 0.6096625328063965, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in een kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ae6b389dcdfb6cce0b41b929271ef7d59e7ac456a2116122b73596159c4f0fb9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "af2a5e9634d3e13b3f3a8456a88ab223014d2db0fa7de5897fb78dca0b256cd6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80e659de1740aa77c901f088df5ad28f1e4d57edbaee34b089decb5a738bd53c", "generation_seconds_batch": 17.484357842011377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3538389205932617, "target_z": 0.36718753580808244, "ce": 6.419747352600098, "genuineness": 4.325342178344727, "identity": 0.47174471616744995, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/af2a5e9634d3e13b3f3a8456a88ab223014d2db0fa7de5897fb78dca0b256cd6.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": "af3f70602e1c03992975730b8c66131e1fe57474cbb4302d7d7a3bc1894386e4", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e20bc8c7db23a4c377f8094382514e9177b56b97dca0c8bb6ecd532ffde33570", "generation_seconds_batch": 18.891724584042095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7175546884536743, "target_z": -0.06982418517992273, "ce": 6.252951145172119, "genuineness": 4.8588056564331055, "identity": 0.6928434371948242, "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_TENS_reg_low/af3f70602e1c03992975730b8c66131e1fe57474cbb4302d7d7a3bc1894386e4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "afda17a34ca8f046dfd8d32a7a6106ad4ec19adc168e49167f72c1bca177374e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbb7b82d42f0bf4ab4ece44685dd3c7e085bdb5024eb81a5a5aafb509a433ab8", "generation_seconds_batch": 22.229551516007632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6570031642913818, "target_z": 0.0029296740719690893, "ce": 6.0305562019348145, "genuineness": 4.494794845581055, "identity": 0.5056243538856506, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/afda17a34ca8f046dfd8d32a7a6106ad4ec19adc168e49167f72c1bca177374e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "afdc5e44b3abc0e9d2801122cb838c017eae3c2f240bef98b029ed0187cbf053", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "130aa57033c09b30c5aba15759fe98c63ef3090f141e1d36935be0fedb127198", "generation_seconds_batch": 16.57739093794953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6862629652023315, "target_z": -0.032226558023989695, "ce": 5.780362129211426, "genuineness": 4.281409740447998, "identity": 0.267366498708725, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/afdc5e44b3abc0e9d2801122cb838c017eae3c2f240bef98b029ed0187cbf053.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "b07a1ef2e605c502a08bed4549eaa27d9f5f74bf295b4eb0d3f89fb76bd5be36", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "38bbdc102c3a766736209b31096e8c7f81e00ac7ddbca761e605d0ce6592615e", "generation_seconds_batch": 18.61482088593766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6494849920272827, "target_z": 0.011962907410038639, "ce": 6.221180438995361, "genuineness": 4.281409740447998, "identity": 0.6553176045417786, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b07a1ef2e605c502a08bed4549eaa27d9f5f74bf295b4eb0d3f89fb76bd5be36.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "b07ade746c7ecd5155548dfa23db2619692db0f0629d9759f4c67bcb4213887e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7f8c66f58839d9e2e0da1783a776bd359bb6f16c6af00920478d257085bcfef", "generation_seconds_batch": 25.467353142099455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1100071668624878, "target_z": 0.6601562320959588, "ce": 5.645336627960205, "genuineness": 4.570107460021973, "identity": 0.5548534393310547, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b07ade746c7ecd5155548dfa23db2619692db0f0629d9759f4c67bcb4213887e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "b09ccbaaa4adb25c1dae43ff584adf981c938d433225179e8e728fa78237c84a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d1f0b5a04f6a13badf21ea064a4d34d8bfe55c251e0ce7e922c25344e06b86f7", "generation_seconds_batch": 18.16572177200578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.586698293685913, "target_z": 0.08740237284406698, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.5565789937973022, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b09ccbaaa4adb25c1dae43ff584adf981c938d433225179e8e728fa78237c84a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "b0c6f1e556437e0f11e06f4db68850f876389a4d78f907941de6faf539ff23ea", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bb2f9ea173c5e321b53e61a7a7ac61c4061f0e8f8edf724053bd6c5788a420e", "generation_seconds_batch": 24.330260206013918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.059325695037842, "target_z": -0.4804688395202061, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.6549757719039917, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b0c6f1e556437e0f11e06f4db68850f876389a4d78f907941de6faf539ff23ea.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "b18d9a9ebabefc3b7c0df1542d7959ebdd22f02cc55df379f0e3480e5915d454", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "456f958fb262de7e2a59cb0e2f8ec1091465023cf412ac6508cf9a877a12d1bb", "generation_seconds_batch": 17.038432899978943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5716620683670044, "target_z": 0.10546869628787636, "ce": 6.062326908111572, "genuineness": 4.174716949462891, "identity": 0.3200633227825165, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b18d9a9ebabefc3b7c0df1542d7959ebdd22f02cc55df379f0e3480e5915d454.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "b18fc03d599c99bc06b2bbd932af8b907b53d944202b858168169dca407321e3", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0b0833c01435b19f36a7c2b16a0e18222cdcec204c3ed7b6f49ef72406de80e7", "generation_seconds_batch": 18.05389176099561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.508265733718872, "target_z": 0.18164065185606182, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.6586721539497375, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b18fc03d599c99bc06b2bbd932af8b907b53d944202b858168169dca407321e3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "b1a01f19d080492c58b4fb638218991f27c88889bd980663f905339429874828", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83b398f7d98ff0cfc17ce8fe1e8fb9b017b774dd938c0e877e9ad859f74d7063", "generation_seconds_batch": 22.347357863909565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7699785232543945, "target_z": -0.13281246419191758, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.5462491512298584, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b1a01f19d080492c58b4fb638218991f27c88889bd980663f905339429874828.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "b2854d33d7420a5480a32f018eaf3b02f42f45ab34780e6a902c69bdc3d723e2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24f8ac4b9fc1bc1b3c9c3cf7d4162a05261f91afc9311e0061649ebc39e25b8d", "generation_seconds_batch": 23.565933825913817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6854501962661743, "target_z": -0.03125, "ce": 5.919358730316162, "genuineness": 4.808597087860107, "identity": 0.6983844041824341, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b2854d33d7420a5480a32f018eaf3b02f42f45ab34780e6a902c69bdc3d723e2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "b2905e04d405d0800ab45bff8097b15d6fab1017e3dc36e8bc69da07ac6d3dba", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb2edc79437c2ccb0db25c2c7cb971ed1308dcb36de4ddaea3e1778c1c458623", "generation_seconds_batch": 16.541026238002814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6143325567245483, "target_z": 0.05419925679608758, "ce": 6.475345611572266, "genuineness": 4.3881025314331055, "identity": 0.47844526171684265, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b2905e04d405d0800ab45bff8097b15d6fab1017e3dc36e8bc69da07ac6d3dba.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "b307d271fd7f23e88b0d67922156747523cae16a205eae271a2cf222eb3ec538", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71066a29642a03b63d4bb218955d586633ff310b019466fae5ad5e172c9897d5", "generation_seconds_batch": 26.697074832045473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5334616899490356, "target_z": 0.1513672098800515, "ce": 6.038498878479004, "genuineness": 4.519899368286133, "identity": 0.532043993473053, "hypothesis": "Y he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b307d271fd7f23e88b0d67922156747523cae16a205eae271a2cf222eb3ec538.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "b336d32d39fb01de68499b8d7e9abd42a51e03258f6b6c4f736045b0303439e2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2157b9791ec22f9e2828d1aa4d78fbad5e6a72c16670df05a35c5726022e2902", "generation_seconds_batch": 24.330260206013918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": -0.644531160479794, "ce": 6.09409761428833, "genuineness": 4.394378662109375, "identity": 0.47341132164001465, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in komen.", "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_TENS_reg_low/b336d32d39fb01de68499b8d7e9abd42a51e03258f6b6c4f736045b0303439e2.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": "b369113ded242a974aa17b145adaa4fa1e250b246e76be8a4dc1a9f68ea2d254", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I wrote the number on the envelope.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34e02b4d8406922725bc6fdfc21e4d00c1e4bd37896a9812fb3bf8b09a9ea734", "generation_seconds_batch": 15.864845247939229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.810617208480835, "target_z": -0.18164065185606182, "ce": 6.4912309646606445, "genuineness": 4.432034492492676, "identity": 0.26545605063438416, "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_TENS_reg_low/b369113ded242a974aa17b145adaa4fa1e250b246e76be8a4dc1a9f68ea2d254.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "b3734f41af1071304d6da4afc434725f1615547f33180b0cd366183a8347c480", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22c4bc08c8459a21c604229a9ea3d38a5922dca4ccda26edff12015f193d9998", "generation_seconds_batch": 16.810488336952403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3616771697998047, "target_z": -0.8437501432323297, "ce": 6.054384231567383, "genuineness": 4.457139015197754, "identity": 0.5339432954788208, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b3734f41af1071304d6da4afc434725f1615547f33180b0cd366183a8347c480.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "b3826b1fc9e7622506ff2c5450cf0c0836ea7f54731fc0d4a97d852059fab6be", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82f02cf896fa4857d622427889027d6add8015dd723288636d625da85238695b", "generation_seconds_batch": 22.30804902408272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8626346588134766, "target_z": -0.24414065185606182, "ce": 6.0305562019348145, "genuineness": 3.9801597595214844, "identity": 0.41655346751213074, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b3826b1fc9e7622506ff2c5450cf0c0836ea7f54731fc0d4a97d852059fab6be.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very relaxed and unstrained. 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": "4f94bc4320a7ba3ebbceb61d5e4c98eb989671993f8203c0dbe93ae86b5f7a8e", "generation_seconds_batch": 18.4237529640086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6405445337295532, "target_z": 0.022705045673925298, "ce": 6.268836498260498, "genuineness": 4.444586753845215, "identity": 0.546574592590332, "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_TENS_reg_low/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "b44a8722d7d82ca0cbb582f1ca617bb96a8c976da1293f81bd2a02c9b2437ece", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b51471b04f1f4e68bd226628b363f6991b9e1e158d45fbc022a4eacaee1597e9", "generation_seconds_batch": 24.367105231038295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5529682636260986, "target_z": 0.1279296740719691, "ce": 6.133810997009277, "genuineness": 4.68307638168335, "identity": 0.7810369729995728, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b44a8722d7d82ca0cbb582f1ca617bb96a8c976da1293f81bd2a02c9b2437ece.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "b46ae4052067c7f9cf5db4553f0e62de9612c93642ece8b95457fcf0435aa76c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57d19afe07a0027108f3fa7da53339cf0a6fc009ced7f3eddf777b3c4dd3e3a6", "generation_seconds_batch": 15.91721858305391, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7821701765060425, "target_z": -0.14746097778409273, "ce": 5.831989288330078, "genuineness": 4.507347106933594, "identity": 0.6293671131134033, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b46ae4052067c7f9cf5db4553f0e62de9612c93642ece8b95457fcf0435aa76c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "b496c3b60f7154dd72f7605213ab87b7aa6cccd743922649c87e1ae29555f9a6", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5431f0887a036f0094a351b353b5d9b5b4a7f26db0dc4d333496e6519724c928", "generation_seconds_batch": 18.28037001506891, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1763648986816406, "target_z": -0.6210937679040412, "ce": 6.157639026641846, "genuineness": 4.168440818786621, "identity": 0.5720943808555603, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b496c3b60f7154dd72f7605213ab87b7aa6cccd743922649c87e1ae29555f9a6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "b4ca20d56d1e22cf73f7f378ccca5b6a6280c9feddc96512df72325a22788f71", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e87f6fd1b46ea5812650018041528c4e686e0f39e58b1997f5b9dc6827b208b8", "generation_seconds_batch": 26.236899528070353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7317782640457153, "target_z": -0.08691409383207213, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.7002753615379333, "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_TENS_reg_low/b4ca20d56d1e22cf73f7f378ccca5b6a6280c9feddc96512df72325a22788f71.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b4e8edc16acb5aed8371b222aa001d5791a95585ea29d2840d957a8700a9d32a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de861bc4ce5502e866a39f1a4cb2588435101a9eb2248379ba85cb3bfcfed2a6", "generation_seconds_batch": 27.577563275001012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2218801975250244, "target_z": -0.6757813037121236, "ce": 5.863759994506836, "genuineness": 4.394378662109375, "identity": 0.5175576210021973, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b4e8edc16acb5aed8371b222aa001d5791a95585ea29d2840d957a8700a9d32a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "b50f0e3cf31f704e4257f25400e70d3f25c49577eaa519f1675d88377a5ee440", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c1d270d3495baf0387a45acd1dfa45f09443a08c0fc0354e530d6133a044c56c", "generation_seconds_batch": 16.24275882903021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8935199975967407, "target_z": -0.28125, "ce": 6.284721851348877, "genuineness": 4.15588903427124, "identity": 0.3678629696369171, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b50f0e3cf31f704e4257f25400e70d3f25c49577eaa519f1675d88377a5ee440.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b620d2dc456c210d0d3edb08e688ee2ebe2d1d265471189157709aebea3a0ec6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33ba50c9336ab0c85046738794333a0f10aca10139ef22b7310c18283c0d3889", "generation_seconds_batch": 22.58143810101319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.377932548522949, "target_z": -0.8632813037121236, "ce": 6.1020402908325195, "genuineness": 4.300237655639648, "identity": 0.6155980229377747, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b620d2dc456c210d0d3edb08e688ee2ebe2d1d265471189157709aebea3a0ec6.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": "b65d4448ef379c134050a2c2c878308f5d1fa5cc1ce5023a00c782e36d54074a", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9a2e8ee81799b9fbac3de3a8907341fb32839c174e0a4870bf2931ca2ce5561", "generation_seconds_batch": 26.802464232081547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6594414710998535, "target_z": -0.0, "ce": 6.268836498260498, "genuineness": 4.65797233581543, "identity": 0.6586637496948242, "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_TENS_reg_low/b65d4448ef379c134050a2c2c878308f5d1fa5cc1ce5023a00c782e36d54074a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "b66e926e6d7b3a5b188ea2d65e543914df3358d393d9cf45e14c20a5ecff328d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fef926581e14c1e5ba571fb1512ebd2f7abe2dff08f9000465f118e087e81371", "generation_seconds_batch": 23.31847012799699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9309074878692627, "target_z": -0.32617181233585574, "ce": 5.732706069946289, "genuineness": 4.168440818786621, "identity": 0.34775736927986145, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b66e926e6d7b3a5b188ea2d65e543914df3358d393d9cf45e14c20a5ecff328d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "b67ab86c4f64d8d403cbd639817e62c39b4648e1d9f6a8c3d93a14c98cbd3bb4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8d329838a1aea6da9251c0a1438bce35e0c830a506fd7bfc1a1261ae252de45", "generation_seconds_batch": 22.87763513904065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.420486330986023, "target_z": 0.2871093481439382, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.27793973684310913, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b67ab86c4f64d8d403cbd639817e62c39b4648e1d9f6a8c3d93a14c98cbd3bb4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "b69fe6607afd95a838fb40f02a75243cc91a49308b5f3faee9acd50b6d0ecfff", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7ce765a4677adb877dc9cfadcd4671ce3c622bb63025f14ccae895ac633f0fe", "generation_seconds_batch": 27.164548998931423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.646437168121338, "target_z": 0.015624928383835142, "ce": 5.847874641418457, "genuineness": 3.8546390533447266, "identity": 0.49155882000923157, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b69fe6607afd95a838fb40f02a75243cc91a49308b5f3faee9acd50b6d0ecfff.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "b6a5862bbf750f26b5bb3fe25b9c5b4ad11af09ed97c343dfb769ae4c45daf12", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e2812fa784e357552fe231f9846defdcc1edd6f8bfb2eda91514c2ee4bfa3f2", "generation_seconds_batch": 25.453008561045863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3180769681930542, "target_z": 0.4101562320959588, "ce": 6.062326908111572, "genuineness": 4.444586753845215, "identity": 0.8516793251037598, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b6a5862bbf750f26b5bb3fe25b9c5b4ad11af09ed97c343dfb769ae4c45daf12.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b6ac8a2c724e2248b874d1d403872274847cb0e5913e688c88f2dd7a2d4be5f2", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e74896d08270697f2e754e2f79bfa88054ad9e8bd4b47c8f24802fe590ee88a7", "generation_seconds_batch": 17.63042104500346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.569746971130371, "target_z": -1.0937501432323298, "ce": 6.1020402908325195, "genuineness": 4.231201171875, "identity": 0.5502908825874329, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b6ac8a2c724e2248b874d1d403872274847cb0e5913e688c88f2dd7a2d4be5f2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "b6b08bba50be7ead60b2fb71348ece6b71e7d027933572fb4edec4b3f2fbb479", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f450da7e767b5ebd3080ae82d9bcbb5a21576181c80b7b0104baa940f8c81cde", "generation_seconds_batch": 22.73483483993914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1587735414505005, "target_z": 0.6015624641919176, "ce": 5.863759994506836, "genuineness": 4.2751336097717285, "identity": 0.8303450345993042, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b6b08bba50be7ead60b2fb71348ece6b71e7d027933572fb4edec4b3f2fbb479.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "b6b4b3872a62c57671c09b2b14ca466b3663bf48ed7b0b6264226685a21dbc56", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "f84a2369c444f3d5dbdc407376010d6edfc11e8a7d4b4b76f5154f86c19934fb", "generation_seconds_batch": 18.32566937804222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4286140203475952, "target_z": 0.2773437679040412, "ce": 6.1020402908325195, "genuineness": 4.507347106933594, "identity": 0.7649470567703247, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b6b4b3872a62c57671c09b2b14ca466b3663bf48ed7b0b6264226685a21dbc56.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "b7027e80a04a2c59c094613e0ef9712cc043fac51dd56fbcb818880c07c3a470", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6051d231948492724a88337669f4004991f5a30a769a16f826c81802387dae14", "generation_seconds_batch": 17.504951542941853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4448695182800293, "target_z": 0.25781246419191756, "ce": 5.95112943649292, "genuineness": 4.281409740447998, "identity": 0.8239376544952393, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b7027e80a04a2c59c094613e0ef9712cc043fac51dd56fbcb818880c07c3a470.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "b7140e2f787319875999787d69a55eadbb9b600529989d5aab2d9774930718f1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0e5781eaebc42044dca2499324fcae46020d442f6ad30d262a8ca082b362354", "generation_seconds_batch": 17.90274157503154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2511401176452637, "target_z": -0.7109376790404122, "ce": 6.046441555023193, "genuineness": 4.331617832183838, "identity": 0.5275443196296692, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b7140e2f787319875999787d69a55eadbb9b600529989d5aab2d9774930718f1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "b7171409438726658fe51cf42a8e2ece7b27416b13ba9b198b150f4bfef3ccce", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33cefc4c2aaf0825e32d666bd521d6e6a47900f145a6e08b15308644ac5ac575", "generation_seconds_batch": 22.85814471996855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6277433633804321, "target_z": 0.038085906167927876, "ce": 6.260893821716309, "genuineness": 4.168440818786621, "identity": 0.27988961338996887, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b7171409438726658fe51cf42a8e2ece7b27416b13ba9b198b150f4bfef3ccce.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b71d79152a51f0912db25d510539d349736eb6b0051ab83fae4ce89abc094f27", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b044d30e774f2f2740486497d8a4c421ffb2ba8d2c2385732f7c5b10ad0a79d", "generation_seconds_batch": 17.286489492980763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.7408344745635986, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b71d79152a51f0912db25d510539d349736eb6b0051ab83fae4ce89abc094f27.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "b7364ea048b3d105541b1087a0c70f2af495111cc61f66c79183c4da05e51d53", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74e1d941e3cca3b23c1604a41bebc09eb5d9a3d809a1d11c1b2b3701b0fb676c", "generation_seconds_batch": 26.98647429398261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8236215114593506, "target_z": -0.19726558023989696, "ce": 6.006728172302246, "genuineness": 4.419482707977295, "identity": 0.6285284757614136, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b7364ea048b3d105541b1087a0c70f2af495111cc61f66c79183c4da05e51d53.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "b7b76ec9199ca991bab2d917729258edff37fa2bf55909eaa91dde849b77519b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2056907087de3a65143c0166fa16e27ceed18fa79e42dbd6e25129d53e56461a", "generation_seconds_batch": 15.988453065045178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1406028270721436, "target_z": -0.5781249283838351, "ce": 6.213237762451172, "genuineness": 4.030367851257324, "identity": 0.35996925830841064, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b7b76ec9199ca991bab2d917729258edff37fa2bf55909eaa91dde849b77519b.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": "b7f5e7f961f0936cb71c1a136945e5c880244bf22aa7cb7a52703c88220ed99e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7f658c37abacdc15692f48160719b9f51788ae1db8e6f19a412d1b4446ccdbe", "generation_seconds_batch": 24.595501671894453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7366548776626587, "target_z": -0.0927734419760103, "ce": 5.935244083404541, "genuineness": 4.7332844734191895, "identity": 0.584631085395813, "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_TENS_reg_low/b7f5e7f961f0936cb71c1a136945e5c880244bf22aa7cb7a52703c88220ed99e.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": "b804c0bad8875a6d440e8f0ce9644dc696c3a5a459683b7f3c96646681936a3b", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e8b125e666603a15bb14ea3b10f3631f45adce8bab601c400aebe6e207949130", "generation_seconds_batch": 18.909186114906333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9097754955291748, "target_z": -0.30078130371212364, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.6879054307937622, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel maar wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b804c0bad8875a6d440e8f0ce9644dc696c3a5a459683b7f3c96646681936a3b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "b88c862ba1645ba35acf79b1777e9ab00aff8911244c0e59acc0b0dd3ca61be2", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "27707a17df351b0462062b44f72da86ba559737b8c4193b18de947229f84f53d", "generation_seconds_batch": 25.694240855053067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2774382829666138, "target_z": 0.458984419760103, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.3599106967449188, "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_TENS_reg_low/b88c862ba1645ba35acf79b1777e9ab00aff8911244c0e59acc0b0dd3ca61be2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "b89dc1250add66ca7ff3991b4827abb4a8d0fc3327caf89e58c94a2d7909330d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "325de642e9ad6feff73f5e0eedd11c519eaea6e151064f84b356beae45edd6fe", "generation_seconds_batch": 27.633057015947998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5915749073028564, "target_z": 0.0815430247001288, "ce": 6.340320587158203, "genuineness": 4.6203155517578125, "identity": 0.5178005695343018, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b89dc1250add66ca7ff3991b4827abb4a8d0fc3327caf89e58c94a2d7909330d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b8b6b444089b0155149daf6f1c1c1375eda4a0cfa764d91e390c18661addfd0c", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8a5382bd1bae4f7c3a3ca65d088540896432ad2ce4fb038b24165c341881c5ae", "generation_seconds_batch": 16.590158174047247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.829833984375, "target_z": -1.4062498567676702, "ce": 6.197352409362793, "genuineness": 4.293961524963379, "identity": 0.7901363968849182, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b8b6b444089b0155149daf6f1c1c1375eda4a0cfa764d91e390c18661addfd0c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "b95b192604b6c64c11f66079bb741ba20ab7dcb524f29f897b57de7118fc2cc9", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9b6dada6979e68934a43ea1e69f41389349fbef11164ff42afaf9fcb4e427003", "generation_seconds_batch": 15.866210230975412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7159291505813599, "target_z": -0.06787106913194334, "ce": 6.308549880981445, "genuineness": 4.306513786315918, "identity": 0.7062214612960815, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b95b192604b6c64c11f66079bb741ba20ab7dcb524f29f897b57de7118fc2cc9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b9b3475b56aeaa58ecfc179a930e95ce6a3008a451d69c945c8008db090bdbe4", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9abcca37c3039ee1aeb4544c733dc4e8b0e25788cf6daf00669e490ad5ff9db", "generation_seconds_batch": 25.429093181039207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6932883262634277, "target_z": -1.2421875358080825, "ce": 6.062326908111572, "genuineness": 4.193545341491699, "identity": 0.7095860838890076, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b9b3475b56aeaa58ecfc179a930e95ce6a3008a451d69c945c8008db090bdbe4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "b9b5f0d3783ed011026c113b42436e3ddf9d631c23d822b8e27fe7e4d1207bb7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e58a90b735c0e730ffd867a535a47a9bdf51198846658d7284490b53dd346cf2", "generation_seconds_batch": 24.11973971198313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9195287227630615, "target_z": -0.3125, "ce": 6.181467056274414, "genuineness": 4.519899368286133, "identity": 0.6489644646644592, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/b9b5f0d3783ed011026c113b42436e3ddf9d631c23d822b8e27fe7e4d1207bb7.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": "ba09e0715e398cafe0726d07f62459bc995d6f8c13950c28ecc3886bac42f140", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "a8e6e8c407fd46cdd48ebdda95e504764134f9734df9635624fc3dee9ae0c098", "generation_seconds_batch": 18.859499237965792, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9130265712738037, "target_z": -0.30468753580808244, "ce": 5.911416053771973, "genuineness": 4.645420074462891, "identity": 0.7956797480583191, "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_TENS_reg_low/ba09e0715e398cafe0726d07f62459bc995d6f8c13950c28ecc3886bac42f140.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very relaxed and unstrained. 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": "64086377bbc10097e37e7b6e1df6ede1659d458ceaad405a5c87b94763a519e5", "generation_seconds_batch": 18.10564396996051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6673660278320312, "target_z": -0.009521512350064397, "ce": 6.221180438995361, "genuineness": 4.6203155517578125, "identity": 0.5728378295898438, "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_TENS_reg_low/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "ba838269384df4aa53345ebc181873c118525f9a8629796bc5638ec0f5fec5fb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "2143e87cba368d009e6239dbaf7c782c1fea2c40e766ad7f952ac7cc334fdb5a", "generation_seconds_batch": 19.09389638202265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.192620277404785, "target_z": -0.6406249283838351, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.5063329339027405, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ba838269384df4aa53345ebc181873c118525f9a8629796bc5638ec0f5fec5fb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "ba8d40b3d9c1796fb8e061ceed0a15b09ab30333e52465a3cdb51b10b45b7c60", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "39b8a951d88c96059dbd2960ff626ab78b6b1e29caf2437100caaec2b6ca3492", "generation_seconds_batch": 19.422527996939607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8122427463531494, "target_z": -0.18359376790404122, "ce": 6.435632705688477, "genuineness": 4.425758361816406, "identity": 0.5819766521453857, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ba8d40b3d9c1796fb8e061ceed0a15b09ab30333e52465a3cdb51b10b45b7c60.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "baafb85578a3f4afc5e8c515b18f29a1d8d69309b7d45d6228fe84cb6c039d1a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "32dfec36371455ebd68a208baaedbaea05decc5d5c5d1e9daf7661542ecfd211", "generation_seconds_batch": 19.423177074058913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0755810737609863, "target_z": -0.5, "ce": 6.078212261199951, "genuineness": 4.362998008728027, "identity": 0.7454221844673157, "hypothesis": "Ik 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_TENS_reg_low/baafb85578a3f4afc5e8c515b18f29a1d8d69309b7d45d6228fe84cb6c039d1a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "babaff53bc9bef5342f6032cc0f1afa4f63e1435dab5f3f27583b7ca60f14290", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1b87800ea0d3e7fd52b865f9e907a087e29cd8c59ef194f62985c4b413a0ffaf", "generation_seconds_batch": 17.10025278897956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5960452556610107, "target_z": 0.07617181233585575, "ce": 6.427689552307129, "genuineness": 4.519899368286133, "identity": 0.6879405379295349, "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_TENS_reg_low/babaff53bc9bef5342f6032cc0f1afa4f63e1435dab5f3f27583b7ca60f14290.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "bae02d40c958768eae7822c3b5188958d87e3b079340190e55cc77e471b1dd51", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20b1844841875ad9711abd9528e2a50154cdbadc7896c7b4481b79e09ae8acd7", "generation_seconds_batch": 27.552371879923157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.366843342781067, "target_z": 0.35156246419191756, "ce": 6.24500846862793, "genuineness": 4.306513786315918, "identity": 0.368072509765625, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bae02d40c958768eae7822c3b5188958d87e3b079340190e55cc77e471b1dd51.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": "bb0d5649bdff25af38105f689c9f1a5c43499928e87e48f042a0f4adcfc61fce", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6dddb2f9f01cfcc38a257a07f593b5222b177aed208e813315e159090ee6d067", "generation_seconds_batch": 26.570803523994982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5440278053283691, "target_z": 0.13867181233585574, "ce": 6.062326908111572, "genuineness": 4.607763767242432, "identity": 0.5690192580223083, "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_TENS_reg_low/bb0d5649bdff25af38105f689c9f1a5c43499928e87e48f042a0f4adcfc61fce.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "bc319450410c3077c0f2089e660fd577879b6349638017272b993aa8fe52d577", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a363c7d2f53c1ed44a8b04942edb3158b9191358c882b832c961dbd476df5d1e", "generation_seconds_batch": 18.494608669891022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0950875282287598, "target_z": -0.5234373925757527, "ce": 6.109982967376709, "genuineness": 4.457139015197754, "identity": 0.6894887685775757, "hypothesis": "De door staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bc319450410c3077c0f2089e660fd577879b6349638017272b993aa8fe52d577.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": "bc467bfa648abb8134db194930fde6033ebdd903ad500c5628cefd9d2df9e38e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c54a7338e4da188e2c2d3a41956e998a93f1f9cbd8b3c00bbed91becf05769f", "generation_seconds_batch": 19.85007185896393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4838825464248657, "target_z": 0.21093753580808242, "ce": 6.260893821716309, "genuineness": 4.570107460021973, "identity": 0.624082624912262, "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_TENS_reg_low/bc467bfa648abb8134db194930fde6033ebdd903ad500c5628cefd9d2df9e38e.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": "bc6343982d686ee77bd130d51db6c2ba49658991e1106df5de9727d5ec6bad07", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1e64a74efaed9ee1e47b19dea04a6a12ad44060b81e7dd8378f60dc1e240987", "generation_seconds_batch": 19.92553039395716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0625767707824707, "target_z": -0.48437507161616483, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.6609678268432617, "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_TENS_reg_low/bc6343982d686ee77bd130d51db6c2ba49658991e1106df5de9727d5ec6bad07.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": "bcea0da8a90a28609bb8f525d9a4bb224db5a6d8273178eb3940d617a6a584ed", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "b2101469e96c6aa7341b8a1b905d0b75191e093361acac7320df813ae7ab6da3", "generation_seconds_batch": 20.661038902006112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.481444239616394, "target_z": 0.2138672098800515, "ce": 5.982900142669678, "genuineness": 4.645420074462891, "identity": 0.446402370929718, "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_TENS_reg_low/bcea0da8a90a28609bb8f525d9a4bb224db5a6d8273178eb3940d617a6a584ed.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "bd3164f16a53d591ccca270af3bbf5b111fd0c3fc4febb8873972425a3d6dacb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c95fbd84cbe4988acfd93e3744b9ecb55d0ba02f234737d6100991806940dc33", "generation_seconds_batch": 23.082437557983212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5659725666046143, "target_z": 0.11230474568813395, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.5835309624671936, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bd3164f16a53d591ccca270af3bbf5b111fd0c3fc4febb8873972425a3d6dacb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "bd380d485548ddf03ebfd973d7af686ce587e929d96b4b5cf887612fcee3db2d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bcd3e0882be5dc45a1a1ae0f2216db6014f13124e7750fb9e1ddfcd19af1bcb8", "generation_seconds_batch": 19.296586651937105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7764806747436523, "target_z": -0.14062492838383514, "ce": 6.173524379730225, "genuineness": 4.18099308013916, "identity": 0.16140688955783844, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bd380d485548ddf03ebfd973d7af686ce587e929d96b4b5cf887612fcee3db2d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "bdf194a0727a96d57e873a7fd8473740173dfc4802463b5df4a71e051c317d02", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4ecd780b47a6d9da974b5498e42ed37d4c91af7d4f9049b0b45f6c53d448197", "generation_seconds_batch": 22.407986855017953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7699785232543945, "target_z": -0.13281246419191758, "ce": 6.0146708488464355, "genuineness": 4.250029563903809, "identity": 0.46629083156585693, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bdf194a0727a96d57e873a7fd8473740173dfc4802463b5df4a71e051c317d02.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "be09cc80a429cb28276ade24abe4bf57c7351ae060899e8fb46084d4b56a6c4d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d60c67e1415a08ff546e72d835f9ae67c9105115314632467db0f2333812c775", "generation_seconds_batch": 25.597969933995046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5675981044769287, "target_z": 0.11035162964015455, "ce": 6.157639026641846, "genuineness": 4.356721878051758, "identity": 0.6822217702865601, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/be09cc80a429cb28276ade24abe4bf57c7351ae060899e8fb46084d4b56a6c4d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "beaee5845779e3ecb6825b75141abca22dec8db1b7eaedcb6a763fb860a8b4fe", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "057dce7bfa52d63b22376f263ebcfa8f7d95e92f49e397de499e484d06004e3d", "generation_seconds_batch": 19.03227954893373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.550529956817627, "target_z": 0.13085934814393818, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.35569241642951965, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/beaee5845779e3ecb6825b75141abca22dec8db1b7eaedcb6a763fb860a8b4fe.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "beb8edcda32feae37dbb5c2f2fe3cf88af0b28340308568749babbff72907ca3", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f05fa21ad35083a169379cfe9cfca36f95aff8b8436bc8726b264acae92842f5", "generation_seconds_batch": 19.189993995940313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8723878860473633, "target_z": -0.2558593481439382, "ce": 6.213237762451172, "genuineness": 4.099404335021973, "identity": 0.3006956875324249, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/beb8edcda32feae37dbb5c2f2fe3cf88af0b28340308568749babbff72907ca3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "bf275b7d5720c5c9abb68aadeaf965a6d64f0846ff102232328123d1ef242b56", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b67ea7d7208b2f28886dc304c80b7703709a1d00e0aca418053d7cc000a2d6d9", "generation_seconds_batch": 25.06517980399076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5619087219238281, "target_z": 0.11718753580808243, "ce": 5.804190158843994, "genuineness": 4.162164688110352, "identity": 0.43035686016082764, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bf275b7d5720c5c9abb68aadeaf965a6d64f0846ff102232328123d1ef242b56.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": "bf3075e34d8cf07e03e4f7cf57f7a1b99188bf1b816129603ee8f9fbbe893229", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3b606af868de82a963c216d22ce19196adef7cd887945cd3edcbddc0084c518", "generation_seconds_batch": 27.31595028808806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0950875282287598, "target_z": -0.5234373925757527, "ce": 6.395919322967529, "genuineness": 4.231201171875, "identity": 0.2828964293003082, "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_TENS_reg_low/bf3075e34d8cf07e03e4f7cf57f7a1b99188bf1b816129603ee8f9fbbe893229.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "bf4132cf653ee3650f4002312f5fa4aa24b881419352b10d3e167c1f851b4b50", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8514b18dcca6988676f763fa63cc9c1c22cd36552338756670808bf967d2bae", "generation_seconds_batch": 22.336135156918317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5521554946899414, "target_z": 0.12890623209595878, "ce": 6.252951145172119, "genuineness": 4.808597087860107, "identity": 0.714336633682251, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bf4132cf653ee3650f4002312f5fa4aa24b881419352b10d3e167c1f851b4b50.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "bf5b6e19c959828012b1b57123e013ec1e961388389669c2cefc18eed881c998", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4408bc1a2090217bde119642ec30a735c51cf00d291f2e93319eab24b5415689", "generation_seconds_batch": 26.325407224008814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3226640224456787, "target_z": -0.7968750716161649, "ce": 5.967014789581299, "genuineness": 4.281409740447998, "identity": 0.6892167329788208, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bf5b6e19c959828012b1b57123e013ec1e961388389669c2cefc18eed881c998.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "bfbf75b7d3c0283343dd9a76c04b750223e2a0c24caf8f25cfb68b8017858bb0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aabef57768b15cc1b671115ead386b7aeec7729da5d88bda669a524ccd2163ec", "generation_seconds_batch": 22.75438572803978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.47087824344635, "target_z": 0.22656246419191758, "ce": 6.41180419921875, "genuineness": 4.325342178344727, "identity": 0.6078448295593262, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bfbf75b7d3c0283343dd9a76c04b750223e2a0c24caf8f25cfb68b8017858bb0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "bfcaef2bf16e69dbafd0a14687faceac16d66d9e547a95805a57ef8f161b8518", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ae61dc848fb59bfc26867c78f4fcefc5eb38f7f1a120cc96a8828096ab02020", "generation_seconds_batch": 24.343175414949656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4042308330535889, "target_z": 0.3066406518560618, "ce": 6.0146708488464355, "genuineness": 4.570107460021973, "identity": 0.7634621858596802, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/bfcaef2bf16e69dbafd0a14687faceac16d66d9e547a95805a57ef8f161b8518.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": "c030d933bc6c3e5dc61846c5fa167d204533cf64e60489052470f4ced975ca29", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "d040db5a8ef836cb5c95b2e3e369cfc6edfee6bb3286276a830a908142d45a37", "generation_seconds_batch": 21.55353422590997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5915749073028564, "target_z": 0.0815430247001288, "ce": 6.427689552307129, "genuineness": 4.607763767242432, "identity": 0.30763739347457886, "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_TENS_reg_low/c030d933bc6c3e5dc61846c5fa167d204533cf64e60489052470f4ced975ca29.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "c0331f5737031f639d25e4cadd73fdab67ff986da218183176c5b3894f4e3ff1", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "43019d6b6ddbd9bbca5a8f588349757fbb14f0d8b1a3755f5d270387a481ecc5", "generation_seconds_batch": 18.05389176099561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.661778211593628, "target_z": -0.002807640127052806, "ce": 5.998785495758057, "genuineness": 4.331617832183838, "identity": 0.5019865036010742, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c0331f5737031f639d25e4cadd73fdab67ff986da218183176c5b3894f4e3ff1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c0869467ccd22cc277116a7dfeb101c6394d978834f58f546df3ba770edfcffb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9109aa7c8f7e0dd080c33fd7784e944b05c7b9d8ed6887e4354b00ecbe0acdcb", "generation_seconds_batch": 19.063538464950398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8723878860473633, "target_z": -0.2558593481439382, "ce": 6.268836498260498, "genuineness": 3.8295347690582275, "identity": 0.08947107195854187, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c0869467ccd22cc277116a7dfeb101c6394d978834f58f546df3ba770edfcffb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "c09eb50330d9e4efe97849db6f7cdb630f4aed8b0cc13c0c041e303418cf1688", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93e3f4a49ee162471bc5e7638fa61113cfcf44e69536367b81b851aa27fe6e06", "generation_seconds_batch": 23.11839755496476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5334616899490356, "target_z": 0.1513672098800515, "ce": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.6679506301879883, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c09eb50330d9e4efe97849db6f7cdb630f4aed8b0cc13c0c041e303418cf1688.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": "c0c85995e9b7ad5c7d3f86e802d24d6996924b0043a6b02b19f4ea8c5116983d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2a89654c6a3a20925f19e816a969ac70502803950ece74178ecf9b3603c10e84", "generation_seconds_batch": 18.283076160936616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.305072546005249, "target_z": 0.42578130371212364, "ce": 5.982900142669678, "genuineness": 4.758388519287109, "identity": 0.594088613986969, "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_TENS_reg_low/c0c85995e9b7ad5c7d3f86e802d24d6996924b0043a6b02b19f4ea8c5116983d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c101702b8da5436697584784e77c4160fbcb7617bd2a71d1845a7d26d92785a6", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "cb5a1b2e5af94746d81e8a0327894e2dd9ed760bfac8aca2e08e80f4ea551633", "generation_seconds_batch": 19.42186652997043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8057405948638916, "target_z": -0.17578130371212364, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.3486015498638153, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c101702b8da5436697584784e77c4160fbcb7617bd2a71d1845a7d26d92785a6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "c1057ef0c9ca6645c28c98aec3a17e919bd7172ff65df3271575261c70384e75", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "163ab08360587ab4f20bf62c35d17b9548f64fb04a81c9d8c46a0433a0c567b2", "generation_seconds_batch": 18.28037001506891, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.45920991897583, "target_z": -0.9609376790404122, "ce": 6.141753673553467, "genuineness": 4.306513786315918, "identity": 0.6089505553245544, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c1057ef0c9ca6645c28c98aec3a17e919bd7172ff65df3271575261c70384e75.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "c174d0858260b763d4e77f0caa25967a0835ea374f2adec9616b246822097542", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8076f0deb14219faf907d902b3b3e0d9e24095a73456e2d691572ce07b47db1", "generation_seconds_batch": 21.44056558900047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.508265733718872, "target_z": 0.18164065185606182, "ce": 6.1894097328186035, "genuineness": 4.545003414154053, "identity": 0.7836650609970093, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c174d0858260b763d4e77f0caa25967a0835ea374f2adec9616b246822097542.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "c2030494204ef50d8920a91ce9435e94824db6e92a4c527cfa57d941e66d3832", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e7cea0965f68f5df4a12dc096d7a8098d9a7bd3d7acc51cbef856a44dcfd191", "generation_seconds_batch": 25.230264256009832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6659436225891113, "target_z": -0.007812464191917571, "ce": 5.669164657592773, "genuineness": 3.8797430992126465, "identity": 0.3930531442165375, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c2030494204ef50d8920a91ce9435e94824db6e92a4c527cfa57d941e66d3832.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "c2270eda01d7c61c4aa7bebd6e778b0a44ef9a84d98c09ddd5f9fd76943d6cbe", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "aa2fafcdeb4bc1b5bf6abfac1ba6ce3ac5ebadeac4c57b133d71f6317a9d08b4", "generation_seconds_batch": 17.016220916062593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.807366132736206, "target_z": -0.17773441976010304, "ce": 6.197352409362793, "genuineness": 4.250029563903809, "identity": 0.45387113094329834, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c2270eda01d7c61c4aa7bebd6e778b0a44ef9a84d98c09ddd5f9fd76943d6cbe.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": "c267716d6169fc185ea547c940254ff2fd6d9104219aaabefc5c50df70b90f85", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "c9840f40a97a4c61d09c190d690405b3daf014f93248a5c1e90c984c5db298c5", "generation_seconds_batch": 17.95438067591749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5927940607070923, "target_z": 0.08007818766414425, "ce": 6.41180419921875, "genuineness": 4.532451152801514, "identity": 0.3733974099159241, "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_TENS_reg_low/c267716d6169fc185ea547c940254ff2fd6d9104219aaabefc5c50df70b90f85.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "c272bc06a47530258cdc777f829c29261cf18d6fef48c8d38c707e858eca9831", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0c55f72ad1e5ad0cfbfd23db401fee9e379efadce9aad86bc59c20853328963", "generation_seconds_batch": 24.87286729400512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.426698923110962, "target_z": -0.9218750716161649, "ce": 6.038498878479004, "genuineness": 4.381826400756836, "identity": 0.7976252436637878, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c272bc06a47530258cdc777f829c29261cf18d6fef48c8d38c707e858eca9831.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c2f087e4531f1963785a86aec24131affd390ab66d6c6005b49bcd02d2cb7378", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7b948dfaf2c9dccf519ce1b34c761fd59033077abd4e60de638f513e5a1d6b7", "generation_seconds_batch": 24.037384828086942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7959872484207153, "target_z": -0.16406246419191758, "ce": 6.292664527893066, "genuineness": 4.243753433227539, "identity": 0.2763933539390564, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c2f087e4531f1963785a86aec24131affd390ab66d6c6005b49bcd02d2cb7378.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very relaxed and unstrained. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5a222ebc2fdf4e418b46b91f7221ade6f72f177a05e408010672f0a7f9800621", "generation_seconds_batch": 17.718867521965876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8626346588134766, "target_z": -0.24414065185606182, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.653586745262146, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c3d8fab505e333c6f86f4518af4bbee5dafe6debaca416e1db5ed742cd8fd296", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97a4f888e4ba5673adf15f2d7621ce1185ad200d75939b25e0a3c738c256d340", "generation_seconds_batch": 17.513587839901447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9878015518188477, "target_z": -0.3945311604797939, "ce": 6.109982967376709, "genuineness": 3.7542223930358887, "identity": 0.1246001273393631, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c3d8fab505e333c6f86f4518af4bbee5dafe6debaca416e1db5ed742cd8fd296.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "c3f5d78cbbe582fddafad5ab1e467a6419eb15c8a9671d5a3f406014ed8d5433", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24b020acf7dd1d49b1f0ec3b60b33772eeca30798d25c1dce661311a0b822673", "generation_seconds_batch": 22.68527123099193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4660015106201172, "target_z": 0.23242195556818546, "ce": 6.1020402908325195, "genuineness": 4.419482707977295, "identity": 0.6949348449707031, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c3f5d78cbbe582fddafad5ab1e467a6419eb15c8a9671d5a3f406014ed8d5433.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "c42f1c17993c92a87c1f80be0b303112806ff56e61a0cb74e1212b7ddbe945f3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d1eb49092975cbb96a5d070c394a3a0cdd27b98197938253022757e4a90937c", "generation_seconds_batch": 26.778572932002135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5724748373031616, "target_z": 0.10449213826388666, "ce": 5.446769714355469, "genuineness": 4.118232727050781, "identity": 0.42421087622642517, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c42f1c17993c92a87c1f80be0b303112806ff56e61a0cb74e1212b7ddbe945f3.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very relaxed and unstrained. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7363bf7ac3f04d379bb2309594cfdd27dec16f68f4d85da35ced3fde9dc4f848", "generation_seconds_batch": 17.750948000932112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0950875282287598, "target_z": -0.5234373925757527, "ce": 6.141753673553467, "genuineness": 4.394378662109375, "identity": 0.517379879951477, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "c49f3250570889d6a5b4712adb4cb9f0a390ecb75a0521ee59c260a38b48dc16", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d5e82c5c339c6f4539d0b18728050a95f4fc44e2fdb8e56b997800831b1802d", "generation_seconds_batch": 24.260291144018993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.636277437210083, "target_z": 0.027832046916036062, "ce": 6.149696350097656, "genuineness": 4.111956596374512, "identity": 0.5591704845428467, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c49f3250570889d6a5b4712adb4cb9f0a390ecb75a0521ee59c260a38b48dc16.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "c507a925e2ed59579829bbe7e528771a09b2395d6ceb5d77f7448d2889eb72b0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52542ce41c5e395db7ba0c695a2e97ff05f5e0be7f4c5c6e62e816b16b1e1b5c", "generation_seconds_batch": 26.778572932002135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.490384817123413, "target_z": 0.20312492838383514, "ce": 5.5539960861206055, "genuineness": 4.2374773025512695, "identity": 0.6231467723846436, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c507a925e2ed59579829bbe7e528771a09b2395d6ceb5d77f7448d2889eb72b0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "c549c5dec230523fb6feedf07ff4995f523f6a02981fe110ae41f9af8959479d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01f4e867c8b54002793bc8f2836ef517beb755f97db08e2268351c67a0c965b6", "generation_seconds_batch": 17.03915092605166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5834472179412842, "target_z": 0.09130860494002575, "ce": 6.483288764953613, "genuineness": 4.381826400756836, "identity": 0.2333797663450241, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c549c5dec230523fb6feedf07ff4995f523f6a02981fe110ae41f9af8959479d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "c564fd1a7d99e68ea172e9a9f5b83b69d211d4e24791c8bc323383a6859909db", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b827ca608f5dc8ae840c7d866a9d998fddd4c6a2d1494fca09ce268aa66854b", "generation_seconds_batch": 23.507540868013166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3099491596221924, "target_z": 0.41992195556818546, "ce": 6.054384231567383, "genuineness": 4.758388519287109, "identity": 0.6966980695724487, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c564fd1a7d99e68ea172e9a9f5b83b69d211d4e24791c8bc323383a6859909db.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": "c57dd68f5eb684ce4f0a7d8800550696b1bd17f2dbf797862adb2453d007fb46", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4ceddd568c982fff3fa4457d7321b052a369f9c902a781b27cfc54c694a701b", "generation_seconds_batch": 19.49101643799804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2934041023254395, "target_z": -0.7617186962878764, "ce": 6.09409761428833, "genuineness": 4.325342178344727, "identity": 0.6569721102714539, "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_TENS_reg_low/c57dd68f5eb684ce4f0a7d8800550696b1bd17f2dbf797862adb2453d007fb46.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "c594bcd2e35bdbbfb066d78143236bdeac65cb8c85cb36af7a9fe6885635f021", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01cf406b871665bf3910e21a2ad3f70a1c7a19a2d2bcd58f7ff3b0cd3d6d854c", "generation_seconds_batch": 17.27087054902222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.230297565460205, "target_z": 0.5156249283838351, "ce": 5.990842819213867, "genuineness": 4.532451152801514, "identity": 0.5603726506233215, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c594bcd2e35bdbbfb066d78143236bdeac65cb8c85cb36af7a9fe6885635f021.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": "c5d68dab9b3a72bcd3b66639a19b837db6ca7cbcc066e24390619efbb35e4e33", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46f52c743c5120959ffff2b06e43747b0295c7efe29c732d063e9c8ff147a288", "generation_seconds_batch": 28.483411198016256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6960161924362183, "target_z": -0.04394525431186606, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.6322261095046997, "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_TENS_reg_low/c5d68dab9b3a72bcd3b66639a19b837db6ca7cbcc066e24390619efbb35e4e33.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": "c61e52de69c5c855639893968cd3ee2b502f2a56ae125a9c41f66a7cff6faeb5", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "915dc350bd44db656a8e81972cebfd4e9e5e25c221345422b4d871bb77c3ac50", "generation_seconds_batch": 19.921026197029278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4318650960922241, "target_z": 0.27343753580808244, "ce": 5.780362129211426, "genuineness": 4.632867813110352, "identity": 0.8496439456939697, "hypothesis": "Die Schlüsse liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c61e52de69c5c855639893968cd3ee2b502f2a56ae125a9c41f66a7cff6faeb5.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": "c6f8dbfd4f5078d15b3a0850ccfe9385683d60db27667fc89395b0bc92367fad", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b642cc73f740ad02efeceaecb7c49944cc401665d2cf2e8d10d1adca64b3be85", "generation_seconds_batch": 26.38442269200459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.383098840713501, "target_z": 0.3320311604797939, "ce": 6.2767791748046875, "genuineness": 4.557555198669434, "identity": 0.7192574739456177, "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_TENS_reg_low/c6f8dbfd4f5078d15b3a0850ccfe9385683d60db27667fc89395b0bc92367fad.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": "c70c8a4e9a54c66ccd360364fd2c74e8ee506a5c661875019e129a837979233b", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "3e6edfbf12c9781425970a3caa999e435046d21175fb71f75f732f06bc3ec511", "generation_seconds_batch": 17.678057114011608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.043070077896118, "target_z": -0.4609373925757527, "ce": 6.459460258483887, "genuineness": 4.406930446624756, "identity": 0.3386659622192383, "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_TENS_reg_low/c70c8a4e9a54c66ccd360364fd2c74e8ee506a5c661875019e129a837979233b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "c70dbbad2eb4db6b5c8b142a56760229b945cf7e927ac444b519d1e0740e4810", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "634d73fe1ddc94ac4c34fa501c267359fb226edd25ab44ed329d76ef419c9e0e", "generation_seconds_batch": 16.91334024700336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6358710527420044, "target_z": 0.02832032592803091, "ce": 6.2767791748046875, "genuineness": 4.532451152801514, "identity": 0.7114723920822144, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c70dbbad2eb4db6b5c8b142a56760229b945cf7e927ac444b519d1e0740e4810.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "c7676d4163f5463ab629ae09d07924b9e7109a1ebf9aa9ae757d1afa02a0a387", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ae48e28c6eab701e6c491a0c87ded0dd502682b587f43716e77a9d0e81977da", "generation_seconds_batch": 23.082437557983212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.366843342781067, "target_z": 0.35156246419191756, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.569510817527771, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c7676d4163f5463ab629ae09d07924b9e7109a1ebf9aa9ae757d1afa02a0a387.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": "c78b343e51e3e63a6f9d4251bde86c7c5ec44d2edecb3f29c6e194b30983907f", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d77df955f7866fe1d9b46a763cad9a60b898139d9ae04fd0fe0e3076bdf95637", "generation_seconds_batch": 20.601640170090832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5781642198562622, "target_z": 0.09765623209595879, "ce": 6.3641486167907715, "genuineness": 4.362998008728027, "identity": 0.2503069341182709, "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_TENS_reg_low/c78b343e51e3e63a6f9d4251bde86c7c5ec44d2edecb3f29c6e194b30983907f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "c7bb844ae7733e1031bc9e46eef55b7b09a18e1f885d8149379d6c4a5f5ef428", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59ad4aa7d03565685261468d698dbde7b81c04dd37d53c7f517f8add4d1fa3a6", "generation_seconds_batch": 22.263895235024393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.312910556793213, "target_z": -0.7851560888636291, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.563125491142273, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c7bb844ae7733e1031bc9e46eef55b7b09a18e1f885d8149379d6c4a5f5ef428.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very relaxed and unstrained. 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": "070fb509627bb0f283e82e9318e8c50c815940852ca5ee15761e7cb41fb87675", "generation_seconds_batch": 17.282651406014338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0658278465270996, "target_z": -0.48828130371212364, "ce": 6.173524379730225, "genuineness": 4.68307638168335, "identity": 0.6239026784896851, "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_TENS_reg_low/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "c823799e90d8f9ffb056854cf0cdddf0177ad996b8f93694e65d1a8f8bc33611", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08b5ffc54e1970a1166dc992be83d638fca7ebbd417a7321be55c25f33f5d4be", "generation_seconds_batch": 18.45995614095591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3879754543304443, "target_z": 0.32617181233585574, "ce": 6.284721851348877, "genuineness": 4.3881025314331055, "identity": 0.31438907980918884, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c823799e90d8f9ffb056854cf0cdddf0177ad996b8f93694e65d1a8f8bc33611.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "c833cbe1362addffeff9aef8a74755f16e9e226b81e5191ca31e73b9dd206394", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1606ef122f2c220858f9612aee24526e3aaf480d85920382595aa9f5e6124e4", "generation_seconds_batch": 24.254216626053676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6647244691848755, "target_z": -0.006347627155933027, "ce": 6.308549880981445, "genuineness": 4.268857479095459, "identity": 0.3649197518825531, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c833cbe1362addffeff9aef8a74755f16e9e226b81e5191ca31e73b9dd206394.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c87509e28dea5fee7a52766e2ab4479aaa5abadbd5ce3cdb698150279931df17", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb8cea5d3eb07919f63bf5f26d27f18b84438aba627265a5d7f44ddbf3fec6f1", "generation_seconds_batch": 25.664999924018048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7472208738327026, "target_z": -0.10546869628787636, "ce": 6.205295085906982, "genuineness": 4.218649387359619, "identity": 0.3670012056827545, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c87509e28dea5fee7a52766e2ab4479aaa5abadbd5ce3cdb698150279931df17.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very relaxed and unstrained. 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": "2c1343c22e3044341cf8692427c7b68e6e6a02a082b857baed91c861f870d33e", "generation_seconds_batch": 18.755763795110397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846084833145142, "target_z": -0.15039065185606182, "ce": 6.324435234069824, "genuineness": 4.582659721374512, "identity": 0.4684150815010071, "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_TENS_reg_low/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "c8bbca17bf17cf7bdda0bc6a1093b8141369a0d113314dfecbbc1bf028c9b806", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e75f47ffca4b1fe63a1f96ad3f5fd4082f5e6e6502298aa49c6f608d32b07644", "generation_seconds_batch": 23.716581071028486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6228667497634888, "target_z": 0.04394525431186606, "ce": 6.459460258483887, "genuineness": 4.243753433227539, "identity": 0.7367229461669922, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c8bbca17bf17cf7bdda0bc6a1093b8141369a0d113314dfecbbc1bf028c9b806.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": "c90132af2a9b894ee2839a5df03c85e6520195499fb623083291d694966477fd", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "834d3955e20291eb21af97ca77e48a39b9a36979ef0f36d5e617769eca0c29f4", "generation_seconds_batch": 15.864832510938868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.662489414215088, "target_z": -0.003662164206126219, "ce": 6.419747352600098, "genuineness": 4.482243061065674, "identity": 0.43391335010528564, "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_TENS_reg_low/c90132af2a9b894ee2839a5df03c85e6520195499fb623083291d694966477fd.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "c92cf86a0bc3d87118b1bd795a857ee5f66fd72664e34e9cd5b5c81550935112", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "d33d67f4369d2b721344b0cdb07b1f0a9dbf88899e57d3c2a1e36f0bbda6496a", "generation_seconds_batch": 19.4367338080192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6214443445205688, "target_z": 0.04565430247001288, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.6445199847221375, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c92cf86a0bc3d87118b1bd795a857ee5f66fd72664e34e9cd5b5c81550935112.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "c9462639c6d45831254d5d2784d5c6c30afd0c67a667508354469b9a511db51c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5f5abbf62c8f9f951489a8096f4fe48343981405311155e9582a444bbef3865", "generation_seconds_batch": 22.73483483993914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3570899963378906, "target_z": 0.36328130371212364, "ce": 5.998785495758057, "genuineness": 4.118232727050781, "identity": 0.6791075468063354, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c9462639c6d45831254d5d2784d5c6c30afd0c67a667508354469b9a511db51c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "c954cb58a472561ef4cdc330877dda1dd1e0a0bc74373f902f8e4b45aded0873", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c76bcfde46ade53f083a7ebebf59e04c0f7242880c41d2d16e0f5a72b6254ff1", "generation_seconds_batch": 15.658181454055011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8666985034942627, "target_z": -0.2490234419760103, "ce": 6.157639026641846, "genuineness": 4.080576419830322, "identity": 0.2804734706878662, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c954cb58a472561ef4cdc330877dda1dd1e0a0bc74373f902f8e4b45aded0873.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "c9df6257188dd6cd1ef699ca9afd1ffc70c7b23453ce4d9d87b8bdc73185374d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be8386a64032a5edcdd9d30f52d00ad1b0f2071e30a4b99bc6c3f63fd9a50cea", "generation_seconds_batch": 27.267675091978163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1165093183517456, "target_z": 0.6523437679040412, "ce": 6.022613525390625, "genuineness": 4.7081804275512695, "identity": 0.7829160690307617, "hypothesis": "J'ai trouvé la petite boite. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble Scalconti.", "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_TENS_reg_low/c9df6257188dd6cd1ef699ca9afd1ffc70c7b23453ce4d9d87b8bdc73185374d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "c9f375ac312c1519a9a3eb6502320436063bfac72dcbfaa2a6bce19985296c8e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ce79ac7f19ccb604ed8ed9b280cc4c5f13411e5ded38f011c1cf942a2d5a3067", "generation_seconds_batch": 16.53960423404351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.306408405303955, "target_z": -0.7773436246717115, "ce": 6.0305562019348145, "genuineness": 4.300237655639648, "identity": 0.7127050161361694, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/c9f375ac312c1519a9a3eb6502320436063bfac72dcbfaa2a6bce19985296c8e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "ca362af58bf6e5420026deac169d5ca31454eec19c99212e062155a08ecdc6cf", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcdd09e9c4dbe303d737e7759b30b5e6fc6e6c3e65d6cf7d769f039d53154afc", "generation_seconds_batch": 20.159247908974066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.270646572113037, "target_z": -0.7343750716161649, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.7590897083282471, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ca362af58bf6e5420026deac169d5ca31454eec19c99212e062155a08ecdc6cf.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "ca641829145ca5e71e30baea00913ab92e162ab0bd8e5c79d69bc89831ed4708", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6df6c33dd7878c2c4033cb3f0c906a3cf8d137dee0941dd8c5b32481dfb20bb", "generation_seconds_batch": 24.261832074960694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3457112312316895, "target_z": 0.3769531160479794, "ce": 5.855817794799805, "genuineness": 4.381826400756836, "identity": 0.6522229909896851, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_TENS_reg_low/ca641829145ca5e71e30baea00913ab92e162ab0bd8e5c79d69bc89831ed4708.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "ca881a6fd837513e04278888393d2f06077fff71c95c8a8952113d470ead9c1e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c795d4c5c2e853455b7b6389f754b7c6ff0dee9c8289cc9155f0ddccd574123", "generation_seconds_batch": 22.68527123099193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4985125064849854, "target_z": 0.19335934814393818, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.666481614112854, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ca881a6fd837513e04278888393d2f06077fff71c95c8a8952113d470ead9c1e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "cbbe08bd87c829f57471df56c47687ff9bb8f8178442f2a2485c92ea22e085ca", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b19eb5c60a155e9359d53970ea5bf9badc8694d925d8a0289e04be7d88ecd3f3", "generation_seconds_batch": 23.92186559503898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.725275993347168, "target_z": -0.07910148640782484, "ce": 6.340320587158203, "genuineness": 4.20609712600708, "identity": 0.6672757863998413, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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_TENS_reg_low/cbbe08bd87c829f57471df56c47687ff9bb8f8178442f2a2485c92ea22e085ca.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": "cbe43acfac2b12e14e587819cbda03042cbbcfd46d5a855c5d6e39f18be70a5b", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90a843dff8ade476a89a4081034d2473db417e4134b79a30af711a27a26740c0", "generation_seconds_batch": 19.899905817001127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4416184425354004, "target_z": 0.26171869628787636, "ce": 5.816103935241699, "genuineness": 4.6956281661987305, "identity": 0.6942415237426758, "hypothesis": "Quelqu'un dut placer la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cbe43acfac2b12e14e587819cbda03042cbbcfd46d5a855c5d6e39f18be70a5b.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": "cca13420f8759f7acf77af97adbd0e7850d9040dc8cb996c56612a1e4f949110", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4564b66698a3002586047e22bdcc87cee9d1d585e6582c94d2549069e47a1ea", "generation_seconds_batch": 24.3374675209634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6066112518310547, "target_z": 0.0634765580239897, "ce": 6.078212261199951, "genuineness": 4.570107460021973, "identity": 0.7794732451438904, "hypothesis": "Te escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cca13420f8759f7acf77af97adbd0e7850d9040dc8cb996c56612a1e4f949110.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "ccb1942b78f496962196d472d937e83034a08c7faeae055e5aee998114253ff3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "18b19a3a591abb94b549d0b46f9506546a32a8999e2cf3bb610c872dda5f3793", "generation_seconds_batch": 26.127805741969496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3489623069763184, "target_z": 0.3730468839520206, "ce": 5.585766792297363, "genuineness": 4.7332844734191895, "identity": 0.5976462960243225, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ccb1942b78f496962196d472d937e83034a08c7faeae055e5aee998114253ff3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "ccb6f3c90d5c7a548c353fcca0c3a376dbc068975c09d03f2d0f43d2c10d8b2f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1732353b8a6a078578791cf48848bb7cd63333d2433aff55f277942ce45b7b4", "generation_seconds_batch": 22.36469458299689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.543215036392212, "target_z": 0.13964837035984545, "ce": 6.070269584655762, "genuineness": 4.519899368286133, "identity": 0.5686864256858826, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ccb6f3c90d5c7a548c353fcca0c3a376dbc068975c09d03f2d0f43d2c10d8b2f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "ccfd66d3bad410ae333babdaf47ca7068b87871947bb150ddda057bc3dd9b7b5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d13e4bbe26cfa4a3f11569e83584f21c18eb4edd56926ed2e4da8e5fd2f38a8", "generation_seconds_batch": 24.36815634998493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3896009922027588, "target_z": 0.32421869628787636, "ce": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.45879167318344116, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ccfd66d3bad410ae333babdaf47ca7068b87871947bb150ddda057bc3dd9b7b5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "cd17f41785a3a5d1abff8183d2b83633c827e9eddc3653710b8d3e7c8266f321", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74778c765d2108ce5c36dfa42082f21a75ea2ca64c2bce058a661786a3b6a803", "generation_seconds_batch": 27.633057015947998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5875110626220703, "target_z": 0.08642581482007727, "ce": 6.324435234069824, "genuineness": 4.118232727050781, "identity": 0.6502699851989746, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cd17f41785a3a5d1abff8183d2b83633c827e9eddc3653710b8d3e7c8266f321.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "cd33b25081e117d1b38132175fa0a076fcf471ca732140f92ec84f2474fb61e6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3d521d5fcbc6c4cb26bda8c7e41bc14986a59d0657047f381125277c5261bf1", "generation_seconds_batch": 23.243780029937625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2660595178604126, "target_z": 0.4726562320959588, "ce": 5.903473377227783, "genuineness": 4.720732688903809, "identity": 0.7937030792236328, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cd33b25081e117d1b38132175fa0a076fcf471ca732140f92ec84f2474fb61e6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "cd96d2601f11840d435a8b347a4d21fce33344ee720a354fe19141a08930b577", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "1b98bff0bc640a37557bd1e69b77093c6b1282c13abdb78186c3a3f32adb7d01", "generation_seconds_batch": 18.33999709400814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9032732248306274, "target_z": -0.29296869628787636, "ce": 6.133810997009277, "genuineness": 4.469690799713135, "identity": 0.5281691551208496, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cd96d2601f11840d435a8b347a4d21fce33344ee720a354fe19141a08930b577.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "ceb000df7b3a29a49c63a71c77d464352eb095f83bba60503d266824a7e94d81", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9abe91410c23687b242e73f49c93a27f9bcd0fd36779ff3053bf48a5bb73a7c", "generation_seconds_batch": 26.72806417301763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8203704357147217, "target_z": -0.19335934814393818, "ce": 5.982900142669678, "genuineness": 4.243753433227539, "identity": 0.538993775844574, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ceb000df7b3a29a49c63a71c77d464352eb095f83bba60503d266824a7e94d81.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "cec89605b6d348fcff6da5d8efa61b71afd7ea03b06fcd4f6debc7cbaead0496", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae6b98fbc6306f903ef76e33b437091398a7a7b321fe3147ff6282ba14b798ca", "generation_seconds_batch": 23.507540868013166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5245212316513062, "target_z": 0.16210934814393818, "ce": 5.808161735534668, "genuineness": 4.507347106933594, "identity": 0.7609751224517822, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cec89605b6d348fcff6da5d8efa61b71afd7ea03b06fcd4f6debc7cbaead0496.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "cee219e5e10b67d5edc052d7c986df253b627ab5f8e60f17dff737477c341094", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1409a94e9174cacf42baabcf50e0de0ca03a83f784cdc79329c3b3c91a0faf1f", "generation_seconds_batch": 25.603557397960685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.598889946937561, "target_z": 0.07275385925189182, "ce": 6.23706579208374, "genuineness": 4.582659721374512, "identity": 0.6100013256072998, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cee219e5e10b67d5edc052d7c986df253b627ab5f8e60f17dff737477c341094.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "cf33eb75858f01b01fdf4b662e58cd0b6903189b1823aa26ef29cf9a447a9605", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7a7ef9a5f825dbc38f7ed6e57a81dde5ea684ea688520eaabfbebbd304eed67", "generation_seconds_batch": 26.66754546505399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6541584730148315, "target_z": 0.006347627155933027, "ce": 6.062326908111572, "genuineness": 4.2374773025512695, "identity": 0.5087788105010986, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cf33eb75858f01b01fdf4b662e58cd0b6903189b1823aa26ef29cf9a447a9605.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very relaxed and unstrained. 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": "684c404fb5fdabd90924fba3dfd666a9c1ee1aa0f5ca7673d59c41e05a5134b7", "generation_seconds_batch": 17.11650668596849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9032732248306274, "target_z": -0.29296869628787636, "ce": 6.316492557525635, "genuineness": 4.300237655639648, "identity": 0.6028403043746948, "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_TENS_reg_low/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "cf9dceb4de2e981f29a40a5b0e5d8c7edc336236dc33ecff17c6223d7c1d15be", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0d065971982d0d442c9c9af6dda61b7ffa40bd6c6668e8e2df42f303c129e315", "generation_seconds_batch": 19.770344026037492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2121267318725586, "target_z": -0.6640623209595878, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.4884917140007019, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cf9dceb4de2e981f29a40a5b0e5d8c7edc336236dc33ecff17c6223d7c1d15be.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "cfb7aae13ee5cedf3168a2e117157fbea3b4c12ef21c28c4696eaac38b9e84aa", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6f592c0371fe3712d11d8cc355b85707f0b8a8554e5b6eb7f3258fd225a7c82", "generation_seconds_batch": 17.631897851941176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7159291505813599, "target_z": -0.06787106913194334, "ce": 6.157639026641846, "genuineness": 4.532451152801514, "identity": 0.6122649908065796, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/cfb7aae13ee5cedf3168a2e117157fbea3b4c12ef21c28c4696eaac38b9e84aa.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d0371c90e46ad8fdb40e8b4ef278e512b9edd99bf90e27fbd2b546adbf8af057", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bffe8c35793b6169ec876889c2e5228e58ea2b87011cbe2b19b8ad1faf5487cf", "generation_seconds_batch": 17.09776693896856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8772644996643066, "target_z": -0.26171869628787636, "ce": 5.752562999725342, "genuineness": 4.406930446624756, "identity": 0.5358596444129944, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d0371c90e46ad8fdb40e8b4ef278e512b9edd99bf90e27fbd2b546adbf8af057.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d087f35c50d32e7b0f031c8b4b4af0c2f081ceee930b5389653181d3e2055a4f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7a3f34e348f0c6b4c4eb3003e72211bc47345ece7eb8d0b4e8f0943e85bd1ddd", "generation_seconds_batch": 18.88197270198725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8301236629486084, "target_z": -0.20507804443181454, "ce": 6.133810997009277, "genuineness": 4.6203155517578125, "identity": 0.6547895073890686, "hypothesis": "Die Tür 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_TENS_reg_low/d087f35c50d32e7b0f031c8b4b4af0c2f081ceee930b5389653181d3e2055a4f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d0ed1687892ee0dbd42cb624ff9470b5c726cf78c58c6eed70b34c26111e2535", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bb8583ccb0e4659be230d79d8c47a29193fe3d77e8239a8178847aa1bc6d41e", "generation_seconds_batch": 24.237824577023275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6793544292449951, "target_z": -0.023925814820077277, "ce": 6.038498878479004, "genuineness": 4.6203155517578125, "identity": 0.6374184489250183, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d0ed1687892ee0dbd42cb624ff9470b5c726cf78c58c6eed70b34c26111e2535.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d11a093b0177a4711e9ec78c86d4df67d5693ab00b352d5271ab069180805a6c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa083e66ef7c86e46be44d33143720fa669f9517daae41a3cbbb55e15257806c", "generation_seconds_batch": 22.899425540002994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6821991205215454, "target_z": -0.027343767904041215, "ce": 6.252951145172119, "genuineness": 4.375550270080566, "identity": 0.6288939714431763, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d11a093b0177a4711e9ec78c86d4df67d5693ab00b352d5271ab069180805a6c.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very relaxed and unstrained. 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": "5fd5ca7565c5d9fd74151443f7606bd9153797f957409d527aa398a192e6119f", "generation_seconds_batch": 17.288285846007057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6903268098831177, "target_z": -0.03710934814393818, "ce": 6.427689552307129, "genuineness": 4.042920112609863, "identity": 0.36693286895751953, "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_TENS_reg_low/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d1518b563fe51ffbb41fe965f1130a8678a73229c0a9f511aa8090da03abbf8e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad127cb77f977c25c3b45c702d7ac74666b9328e2122a792d015c762b3d7bd54", "generation_seconds_batch": 18.20107739896048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7943617105484009, "target_z": -0.16210934814393818, "ce": 6.0305562019348145, "genuineness": 4.482243061065674, "identity": 0.7316889762878418, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d1518b563fe51ffbb41fe965f1130a8678a73229c0a9f511aa8090da03abbf8e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "d165adc8fbcb72305f7574f4f11f62db3dca93456b37cd370467bc1e85665e29", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "680e1adb4a4a215b019d732ae68d4e3469c2f02c5dcba1fd5d84cb9829f1a596", "generation_seconds_batch": 23.15190843003802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5521554946899414, "target_z": 0.12890623209595878, "ce": 6.173524379730225, "genuineness": 4.218649387359619, "identity": 0.31213200092315674, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d165adc8fbcb72305f7574f4f11f62db3dca93456b37cd370467bc1e85665e29.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "d1c07f98eec087a9ef62d5001cd8facd66063bb1bd51d9755b5882ab3a69a75d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a6724f212878e782049756494a805ceac2538ea1570bf71f29aa9c1d24ae735f", "generation_seconds_batch": 16.984286742983386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7037376165390015, "target_z": -0.053222698772097884, "ce": 6.2767791748046875, "genuineness": 4.419482707977295, "identity": 0.6646169424057007, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d1c07f98eec087a9ef62d5001cd8facd66063bb1bd51d9755b5882ab3a69a75d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d1d2eb8ec499716ea8bf09e7490b976268cb63fd240d5f4b03ca55956b81d48a", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "9bd0d6f8d5a73b08deedb6771a9e2127f2e4bc53545f17c343f610bc589c6ecc", "generation_seconds_batch": 18.07121021405328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7183674573898315, "target_z": -0.07080074320391241, "ce": 5.895530700683594, "genuineness": 4.344170093536377, "identity": 0.4968224763870239, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d1d2eb8ec499716ea8bf09e7490b976268cb63fd240d5f4b03ca55956b81d48a.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very relaxed and unstrained. 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": "cb89a042dddaae93d218a9ddfba6d03503f7843bd28ba95c9f3c1d34d6f704ef", "generation_seconds_batch": 19.72736270702444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846084833145142, "target_z": -0.15039065185606182, "ce": 6.173524379730225, "genuineness": 4.519899368286133, "identity": 0.5968960523605347, "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_TENS_reg_low/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.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": "d29834728cf060a1688fe642ee122a980677c3c933e5773b3191be09e287db61", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "4ebb1a48ce9db68f28cb76060bc7079a775c5fe6f4496836afd435010b71b93f", "generation_seconds_batch": 20.627625997993164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4521844387054443, "target_z": 0.2490234419760103, "ce": 5.514282703399658, "genuineness": 4.7458367347717285, "identity": 0.7808165550231934, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d29834728cf060a1688fe642ee122a980677c3c933e5773b3191be09e287db61.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "d299a8735344781550cf7b0a7d0e9d2d1a0b074e013f8d9e1f91ae64d6cb1073", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2057369b6f1e5002046a178d2e00b39ba661a95df4dfe868a3275c9328b241eb", "generation_seconds_batch": 23.067874257918447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6251018047332764, "target_z": 0.04125979136205925, "ce": 6.395919322967529, "genuineness": 4.20609712600708, "identity": 0.701614260673523, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d299a8735344781550cf7b0a7d0e9d2d1a0b074e013f8d9e1f91ae64d6cb1073.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d2cead5164fbcf1680b1417b7d712ef1a5ce724367e2865d7cd41769b1ff5e49", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "7689eab6dfe7a1ac50b6d2189f6b290438da91494856abeba3b0666e64794f7f", "generation_seconds_batch": 18.08223382907454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7740423679351807, "target_z": -0.13769525431186605, "ce": 5.681078910827637, "genuineness": 3.7291181087493896, "identity": 0.5428776741027832, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d2cead5164fbcf1680b1417b7d712ef1a5ce724367e2865d7cd41769b1ff5e49.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "d2d36ed23424d3228c778df2f8ef43be43b1d7cd9b3cffb8836a36f5697dee9e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00a498a4b97c2c0d0eeb6fb71d48049b3b13ae08c2dec3339c91111f84d4c100", "generation_seconds_batch": 25.453008561045863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.123011589050293, "target_z": 0.644531160479794, "ce": 6.078212261199951, "genuineness": 4.582659721374512, "identity": 0.8414994478225708, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d2d36ed23424d3228c778df2f8ef43be43b1d7cd9b3cffb8836a36f5697dee9e.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": "d2e74b2444c9045edde4491c67dda974c982bcc09b157713d3c48ab3679d703e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1505aae005f4afe4f4dd036a1d329f73be1937709d1efae36b35168dd8a133f7", "generation_seconds_batch": 26.06688578496687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3961031436920166, "target_z": 0.3164062320959588, "ce": 6.4515180587768555, "genuineness": 4.494794845581055, "identity": 0.22669269144535065, "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_TENS_reg_low/d2e74b2444c9045edde4491c67dda974c982bcc09b157713d3c48ab3679d703e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "d311facca1dd965b440530b063c41dcf95fd139b3b74f9b0b1691141ff624140", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3654bd9594264ed405e477d1f4b53ce7c6d2a402b7eb98c067c586dd9b7d0e25", "generation_seconds_batch": 24.01814999501221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.883766770362854, "target_z": -0.26953130371212364, "ce": 6.260893821716309, "genuineness": 4.168440818786621, "identity": 0.6359301805496216, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d311facca1dd965b440530b063c41dcf95fd139b3b74f9b0b1691141ff624140.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "d3568872d9e9cb2af717deb3f6107bd0daabf87057ade947e0436bb0f0bc891d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eec15e3e98141a1f63afb2a4860b5eb9ecd6520dd1c7c1decd4979e173149ac7", "generation_seconds_batch": 19.092876061913557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.435116171836853, "target_z": 0.26953130371212364, "ce": 6.252951145172119, "genuineness": 4.6705241203308105, "identity": 0.7940970063209534, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d3568872d9e9cb2af717deb3f6107bd0daabf87057ade947e0436bb0f0bc891d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d3ab0c0fd1569199a49a9ee4a045763a58d99f673be2f14c2b7a1515bb839090", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "088427fa7f2f4ca0a2a77ce98e2c13e60aa6ee873c50a98ba9a34d78b71a55b3", "generation_seconds_batch": 25.95990879496094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9309074878692627, "target_z": -0.32617181233585574, "ce": 6.221180438995361, "genuineness": 4.8588056564331055, "identity": 0.6720569729804993, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d3ab0c0fd1569199a49a9ee4a045763a58d99f673be2f14c2b7a1515bb839090.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d3bb4ebf1cf02ec51d56c41b1dacb5d8783b569e4af70a6f223b63cdda38e09b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "505478e3da581a40fb32cddd38c412600ab5ced8e178507379b2b67911301615", "generation_seconds_batch": 25.997556727030315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4042308330535889, "target_z": 0.3066406518560618, "ce": 5.974957466125488, "genuineness": 4.149612903594971, "identity": 0.5029182434082031, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d3bb4ebf1cf02ec51d56c41b1dacb5d8783b569e4af70a6f223b63cdda38e09b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "d3bb82149dd393f124fa32542352ab00fbc32596cdc8517444ccf0fedda13591", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "787dcaa5724a2baa1fe6073cc3db7a5e1ad9faef349a46c6e51ed05fc2bd4fd9", "generation_seconds_batch": 23.9797532169614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.002431631088257, "target_z": -0.4121094913762679, "ce": 6.141753673553467, "genuineness": 4.362998008728027, "identity": 0.6376997828483582, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d3bb82149dd393f124fa32542352ab00fbc32596cdc8517444ccf0fedda13591.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "d3c744f59988d6f862f2d8dc0e6e79a808856fedd4f18501f79fa779894e28d0", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ef9ec94854f976153639b9bb1886bebc5ed15d4a7bd9fa9fe7607b581e81f739", "generation_seconds_batch": 16.72145465493668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7171483039855957, "target_z": -0.06933590616792787, "ce": 6.340320587158203, "genuineness": 4.645420074462891, "identity": 0.34599626064300537, "hypothesis": "I found the little box. It was beside the old window. We can look inside together", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d3c744f59988d6f862f2d8dc0e6e79a808856fedd4f18501f79fa779894e28d0.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very relaxed and unstrained. 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": "8d3b2f198f993f916b29c1a0ae9ee702186f0faa0ebd56ad285b8ca3cc78cc49", "generation_seconds_batch": 17.60408251208719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.538338303565979, "target_z": 0.14550786173611333, "ce": 6.0305562019348145, "genuineness": 4.494794845581055, "identity": 0.8181911110877991, "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_TENS_reg_low/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d40d87536d7be2932698489491a4cfaad90cbe25dee2bf2d0f3001d0e5b93a6e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8c33d92cdbd249263dda954393cc85271fe3e22fc7023884080d2527ca3eb87c", "generation_seconds_batch": 17.48881007207092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0658278465270996, "target_z": -0.48828130371212364, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.49061593413352966, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d40d87536d7be2932698489491a4cfaad90cbe25dee2bf2d0f3001d0e5b93a6e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d4780235fe00209c3a7be3729b6ea9bf485a134a9f3c60e126a46eaaabd5510e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af413c111ea145e2099e9ed2c22f19fcf577c9c3b1270842885992019fbbae58", "generation_seconds_batch": 23.595418605022132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.807366132736206, "target_z": -0.17773441976010304, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.3208606243133545, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d4780235fe00209c3a7be3729b6ea9bf485a134a9f3c60e126a46eaaabd5510e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d4f7ee6ddf02f9391ddb6f67ae59594906fc18edf024a288de649a4afa6aa561", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47df86a94e349b9cc3df9294315e31e472a7bd41420a90fc0b6e6cb9e8a2f8a5", "generation_seconds_batch": 23.499189599999227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.618802785873413, "target_z": 0.04882818766414425, "ce": 6.2767791748046875, "genuineness": 4.325342178344727, "identity": 0.42061173915863037, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d4f7ee6ddf02f9391ddb6f67ae59594906fc18edf024a288de649a4afa6aa561.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d51b95823ba92664362864c7f4c7864d95dc4285b837c6e0cf7437aa0be8e456", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ece79e1dc3c27bfcabd429954c1f3ad0a6331cb01b759d461ee9691ba453d34", "generation_seconds_batch": 24.126441917032935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.684637427330017, "target_z": -0.030273441976010302, "ce": 5.935244083404541, "genuineness": 4.18726921081543, "identity": 0.6659839153289795, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d51b95823ba92664362864c7f4c7864d95dc4285b837c6e0cf7437aa0be8e456.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "d57c412aa58cb39526492ddc31a8545c3b886cf6c8cd46c4d66606d7eafb043d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78f226c4250c2375c127a0eb7ac26d5472913a02fe3de484865dcbfac7280bcc", "generation_seconds_batch": 25.375774536048993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.887017846107483, "target_z": -0.27343753580808244, "ce": 5.804190158843994, "genuineness": 4.143336772918701, "identity": 0.5987117886543274, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d57c412aa58cb39526492ddc31a8545c3b886cf6c8cd46c4d66606d7eafb043d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "d6523e59e8e80e7020a8d7e9970679c08ad6a613605052a95702fe47f88fd7a8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae0fcd59fa2da404ccc39fb5449c8e81e8747aa83cb513b0589ee50353fa7779", "generation_seconds_batch": 26.69907436799258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5228956937789917, "target_z": 0.16406246419191758, "ce": 6.205295085906982, "genuineness": 4.400654315948486, "identity": 0.7172966003417969, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d6523e59e8e80e7020a8d7e9970679c08ad6a613605052a95702fe47f88fd7a8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "d6eba0636a0fe9f2e9164021d2d484a8f7af924f1945f250fc48a12469b807b0", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c209ffed71090973fc21fc8adcb9ddcbb0649d8b757d94ba98da01243e579870", "generation_seconds_batch": 22.739485797006637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4221118688583374, "target_z": 0.2851562320959588, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.7647730112075806, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d6eba0636a0fe9f2e9164021d2d484a8f7af924f1945f250fc48a12469b807b0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "d7008352ea7fd250243e6cc8ee79c916d2cb506ba5f32d415c9d909b5ffb041a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "289452a085e828b3591214b0328074567543818510fa6597dac7a6160d7f9c19", "generation_seconds_batch": 24.256843673996627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1587735414505005, "target_z": 0.6015624641919176, "ce": 6.0146708488464355, "genuineness": 4.331617832183838, "identity": 0.3978455364704132, "hypothesis": "I found the little box. It was beside the old window. We can look inside", "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_TENS_reg_low/d7008352ea7fd250243e6cc8ee79c916d2cb506ba5f32d415c9d909b5ffb041a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "d774c2c193949501ea7759142eed5bc9d2b81a258c35e0dea0f4704929e184e7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7e74e7d109f4b95020eb07cc144bf40ea36c83050791a7f229b6ca56661b0b4", "generation_seconds_batch": 21.058796579018235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3941879272460938, "target_z": -0.8828124641919176, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.8122900128364563, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d774c2c193949501ea7759142eed5bc9d2b81a258c35e0dea0f4704929e184e7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "d7ad41eb3e09f6baeeda947a353aca8845485eb49c87189a7b766216a7eccf35", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0477d57e3b0ffeee02b8853d257a24f27756679649f8082dfd3ef066db88593", "generation_seconds_batch": 24.70449912897311, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5822280645370483, "target_z": 0.0927734419760103, "ce": 6.2767791748046875, "genuineness": 4.281409740447998, "identity": 0.6140331029891968, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d7ad41eb3e09f6baeeda947a353aca8845485eb49c87189a7b766216a7eccf35.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": "d87777c04a9d98983463a8e7b09b3129b5bd4e1cf32d4f5ee1e1c2e713dfff24", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a60c21e9f0885fd886a105b9225743d14f0d54877033b5dde8b15f135de54b8c", "generation_seconds_batch": 22.808008905034512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5033891201019287, "target_z": 0.1875, "ce": 6.23706579208374, "genuineness": 4.758388519287109, "identity": 0.7627685070037842, "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_TENS_reg_low/d87777c04a9d98983463a8e7b09b3129b5bd4e1cf32d4f5ee1e1c2e713dfff24.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d8b0532e9d0a72eefa20a0f82f932361029c91db078f7bb493100bde0b293c17", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa530411e4be6e7d34447c852811efdd3ef17b26d4ee8d017e8ee67bfe5098be", "generation_seconds_batch": 23.654159380937926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8585708141326904, "target_z": -0.23925786173611333, "ce": 6.3641486167907715, "genuineness": 4.268857479095459, "identity": 0.21557223796844482, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d8b0532e9d0a72eefa20a0f82f932361029c91db078f7bb493100bde0b293c17.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d8ff1ab223d6550c8e1f8d91295cc822f112b91a31664d5912154f250ea6e635", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "448e5fce54733fa6797821b5b278c4a9f7e70683e3d70b1d562ba83b990bc54a", "generation_seconds_batch": 17.950093136983924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9812994003295898, "target_z": -0.38671869628787636, "ce": 6.197352409362793, "genuineness": 3.9173994064331055, "identity": 0.21058356761932373, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d8ff1ab223d6550c8e1f8d91295cc822f112b91a31664d5912154f250ea6e635.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d93cefaf3593f316d1b4f4b7a773f16d6ffdf54be0d4897cbb407f79d10e4967", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9ca963e8548a3a0382f4b78e4016e3250791e04b73f782a6a84dec602b998f2", "generation_seconds_batch": 23.654159380937926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9292819499969482, "target_z": -0.32421869628787636, "ce": 6.268836498260498, "genuineness": 4.331617832183838, "identity": 0.3986744284629822, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d93cefaf3593f316d1b4f4b7a773f16d6ffdf54be0d4897cbb407f79d10e4967.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": "d9487202210e8ecdd9169f070cb20309d98a512afa8e93929f586e61932f4284", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88962b04745d72109c48d33273c5088c802ad6bdb1dcd305c0eee8e17298dccc", "generation_seconds_batch": 20.038881281972863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9146521091461182, "target_z": -0.3066406518560618, "ce": 6.483288764953613, "genuineness": 4.545003414154053, "identity": 0.18328697979450226, "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_TENS_reg_low/d9487202210e8ecdd9169f070cb20309d98a512afa8e93929f586e61932f4284.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "d9541c71128ba1cb1621adda315affa102a30ca82dc9f0716fb947a98bd01472", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d1ffcfdb8bf7006b5962b85794016fc8d9ffbdc8e1a6995b6ce5566243a2402", "generation_seconds_batch": 22.49123009899631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6293689012527466, "target_z": 0.03613279011994848, "ce": 6.284721851348877, "genuineness": 4.432034492492676, "identity": 0.3612014055252075, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d9541c71128ba1cb1621adda315affa102a30ca82dc9f0716fb947a98bd01472.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "d96fc40c298dbcf7d0442d55a1fe472d67085c98abd63d48e217f7cad47130ab", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fe0bd72c566cbeb0ce82f7e26414cfef8eec1b55bcdffb9e7a6cc82540ae38b", "generation_seconds_batch": 24.254216626053676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5456533432006836, "target_z": 0.13671869628787636, "ce": 6.443574905395508, "genuineness": 4.482243061065674, "identity": 0.445766806602478, "hypothesis": "I found the little box. It was beside the old window. We can look inside together", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d96fc40c298dbcf7d0442d55a1fe472d67085c98abd63d48e217f7cad47130ab.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "d97ab249e1acf6b299dfde33842508399447a218a6a8d0df64d58152b888bb56", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21c95a6e9c7394c438a9dbb425c5bd40d71cead99debff5b1cad23bbc88939d9", "generation_seconds_batch": 23.45474422397092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.46518874168396, "target_z": 0.23339851359217517, "ce": 5.530168056488037, "genuineness": 4.356721878051758, "identity": 0.7499419450759888, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d97ab249e1acf6b299dfde33842508399447a218a6a8d0df64d58152b888bb56.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "d9b5905fb082a372995e9162cc931e2709efd9dd7e4cdef76ba88fc0f889e9f5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ceefc0c3420ffc01dbabc433bd008b8b5d7c1e44507899e5998cc8057ebbd10", "generation_seconds_batch": 22.92645551392343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3489623069763184, "target_z": 0.3730468839520206, "ce": 6.340320587158203, "genuineness": 4.300237655639648, "identity": 0.3205726444721222, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d9b5905fb082a372995e9162cc931e2709efd9dd7e4cdef76ba88fc0f889e9f5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "d9c4d9586d01f5e6eb6adfaf056b81ec5f56c2f0cf501f9f987b639e314fad8b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8305354d5c4bc8a3a1e1fec410c27c723d024f5bc073f6333c68e7e1800f679d", "generation_seconds_batch": 18.019026340916753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4091074466705322, "target_z": 0.30078130371212364, "ce": 6.1020402908325195, "genuineness": 4.287685871124268, "identity": 0.555109441280365, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d9c4d9586d01f5e6eb6adfaf056b81ec5f56c2f0cf501f9f987b639e314fad8b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "d9c87ccd538c700c2ce4aa562349a6b54b8bbf6fb510ea6075257e5942789c19", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb9ead87642e3241189fcade358aac05ab3439e30c07aa3b0e68cb4c3d290e5b", "generation_seconds_batch": 24.248063872917555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.691139578819275, "target_z": -0.038085906167927876, "ce": 6.157639026641846, "genuineness": 3.7291181087493896, "identity": 0.34237968921661377, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/d9c87ccd538c700c2ce4aa562349a6b54b8bbf6fb510ea6075257e5942789c19.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": "da1421f25751e924910bfdd648852b39f17ae438a9bc876ee92b02e44c843e8c", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25edbfc7754be951cb5de471081401305ef90c5d91b97816babd7080cc8c284d", "generation_seconds_batch": 22.970506611978635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8293108940124512, "target_z": -0.20410148640782483, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.5934563875198364, "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_TENS_reg_low/da1421f25751e924910bfdd648852b39f17ae438a9bc876ee92b02e44c843e8c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "da1c6dd93a8cfae6156f6e6aee09485dd2887ad3dac80d361805ba90e5e72b7b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af7d76e14875e12ce88a845de549a817225f4f08c56e007092a290ca1e92fdc0", "generation_seconds_batch": 19.062023898935877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7862340211868286, "target_z": -0.15234376790404122, "ce": 6.292664527893066, "genuineness": 4.118232727050781, "identity": 0.5206969380378723, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/da1c6dd93a8cfae6156f6e6aee09485dd2887ad3dac80d361805ba90e5e72b7b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "da2136208d7b173b16d77b4ddd2486519ce43963f9f27396bcb99a904e5f7344", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aceb726a0985e3f3cb65254efd6646fa73a20434b39b68e79cf92f057a405c6c", "generation_seconds_batch": 22.58143810101319, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.199122428894043, "target_z": -0.6484373925757527, "ce": 5.927301406860352, "genuineness": 4.369274139404297, "identity": 0.6141130924224854, "hypothesis": "De door staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/da2136208d7b173b16d77b4ddd2486519ce43963f9f27396bcb99a904e5f7344.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "da2d54cdcc9621057a8d03468aaf18f6e7f15d48d2f6544054aa2f76a4cfc45f", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e1638b6d5e9787f76517f8c7471e059800337c452dfdd1d834c216275b306ada", "generation_seconds_batch": 16.325509530026466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7313718795776367, "target_z": -0.08642581482007727, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.4838654100894928, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/da2d54cdcc9621057a8d03468aaf18f6e7f15d48d2f6544054aa2f76a4cfc45f.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": "da566db9a13f5124df2914629cdc05c9dfb2b7cffbdbb4a8ed26932856badb45", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ed264d0b3049660d65365cc6168934f341df3c1fc2097de6b7002f83a63a6e1", "generation_seconds_batch": 24.09276328701526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7520976066589355, "target_z": -0.11132818766414425, "ce": 6.062326908111572, "genuineness": 4.570107460021973, "identity": 0.6876559257507324, "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_TENS_reg_low/da566db9a13f5124df2914629cdc05c9dfb2b7cffbdbb4a8ed26932856badb45.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "da7c759f4fc6377f7661eb349c53e67e0a684b0203d396519edbee33498fc178", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60c6effec299ea379d847caf63129880cd05d0aac76414427d61beae5407002b", "generation_seconds_batch": 19.41557416692376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.043070077896118, "target_z": -0.4609373925757527, "ce": 5.585766792297363, "genuineness": 4.268857479095459, "identity": 0.4661061763763428, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/da7c759f4fc6377f7661eb349c53e67e0a684b0203d396519edbee33498fc178.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very relaxed and unstrained. 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": "f4a98f5f53edca28df5df75af1ed08914411c2e1815dabbf454d5af24900c5cc", "generation_seconds_batch": 17.68461528304033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0300657749176025, "target_z": -0.44531246419191756, "ce": 6.372091293334961, "genuineness": 4.281409740447998, "identity": 0.30812668800354004, "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_TENS_reg_low/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "db9c9d800d2247bb9af11c90d79d85e94eea47dfb4a5681050a93ac0971427ee", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "114de8690df3c4d5e3a118bf0e71270e02bd58a9c764c36c1135674a511f4b93", "generation_seconds_batch": 18.08223382907454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.554593801498413, "target_z": 0.1259765580239897, "ce": 5.681078910827637, "genuineness": 4.174716949462891, "identity": 0.6217976808547974, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/db9c9d800d2247bb9af11c90d79d85e94eea47dfb4a5681050a93ac0971427ee.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "dbafe8a2f827b94a917cafd7073551f8f6601e1746bbe4432dee8775bd59ec16", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aba3fd2bb7515c1fb2dbc754741f82ed505969635429e8986b49c3193e914605", "generation_seconds_batch": 24.669813812011853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083998441696167, "target_z": 0.6914062320959588, "ce": 6.117925643920898, "genuineness": 4.18726921081543, "identity": 0.7799168825149536, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dbafe8a2f827b94a917cafd7073551f8f6601e1746bbe4432dee8775bd59ec16.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "dbd93a252c03dfd3ba553975c22afde15ce4888b94be47ab60f184473d702f6a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1609b25338c7a5e7adc6addf0304385c071527f109804cde5e3d6f5e1b6a7c8e", "generation_seconds_batch": 17.43291210499592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3375835418701172, "target_z": 0.38671869628787636, "ce": 6.117925643920898, "genuineness": 4.6203155517578125, "identity": 0.8492298126220703, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/dbd93a252c03dfd3ba553975c22afde15ce4888b94be47ab60f184473d702f6a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "dc3e665f3ec12d03a70cbc2d8ed0e1f8f89b197b4df4824c4c603434c9fac93b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c3a1a54e61e9e0e068aa4b187fff617a19f424b86738b1bf0aec3c46308b482", "generation_seconds_batch": 25.53537341102492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1262626647949219, "target_z": 0.6406249283838351, "ce": 5.95112943649292, "genuineness": 4.2374773025512695, "identity": 0.5693091750144958, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dc3e665f3ec12d03a70cbc2d8ed0e1f8f89b197b4df4824c4c603434c9fac93b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "dc456cb071b493b96210c982ec701fe233df93cede4f846dadbe0f9ce8c3bb05", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6d799cfbe85ab81e7946db09d0c292c23678805f1fdfd2e9faf2925686d2f9c", "generation_seconds_batch": 23.31476210197434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8853923082351685, "target_z": -0.271484419760103, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.6443710923194885, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dc456cb071b493b96210c982ec701fe233df93cede4f846dadbe0f9ce8c3bb05.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "dc4d66be3f0c9f0595fc679974fa9c130de3bfece5c49dd9a58989124692b359", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "714996e6c60eefd12ad15734561ea22d6c7b98963cc2dee358119c5086faa5ca", "generation_seconds_batch": 20.040056150988676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9032732248306274, "target_z": -0.29296869628787636, "ce": 6.205295085906982, "genuineness": 4.042920112609863, "identity": 0.36739248037338257, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dc4d66be3f0c9f0595fc679974fa9c130de3bfece5c49dd9a58989124692b359.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "dc5157724dc217c23056d725443fa646c3e25d7be1373dc82cacc3d483a241eb", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5dc92b7b955d3895e9457640cd63c52f766a976629379ddc0389960386a911de", "generation_seconds_batch": 17.887065967079252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.677119255065918, "target_z": -0.021240208637940756, "ce": 6.443574905395508, "genuineness": 4.300237655639648, "identity": 0.413284033536911, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dc5157724dc217c23056d725443fa646c3e25d7be1373dc82cacc3d483a241eb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "dca39c43d79a45c18bb41f2f703f354b35b855f6327a1ed4889a9cfad8e9b1cc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9cdd88ba7cb67a86962130e10f2beed2574a63c03aec94eb36cfa54e7bb0fd4", "generation_seconds_batch": 24.377054914017208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4806314706802368, "target_z": 0.21484376790404122, "ce": 6.292664527893066, "genuineness": 4.268857479095459, "identity": 0.6311440467834473, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinkommen.", "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_TENS_reg_low/dca39c43d79a45c18bb41f2f703f354b35b855f6327a1ed4889a9cfad8e9b1cc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "dd2315a537fd1acc9ed4f4fc26783d668b38552d9a72dee930ecc3fbb07efff3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e128c7198289bf135dd11691a4a0e6ccd589187b2f4a96f444c8b94263cd0d1f", "generation_seconds_batch": 21.940618842025287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5017635822296143, "target_z": 0.1894531160479794, "ce": 6.157639026641846, "genuineness": 4.394378662109375, "identity": 0.6778452396392822, "hypothesis": "J'ai trouvé la pute de boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "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_TENS_reg_low/dd2315a537fd1acc9ed4f4fc26783d668b38552d9a72dee930ecc3fbb07efff3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "dd7010c16ea5997a0876dc4f7f35fde217a9f5e6b6fa14853c10b978b5bea8a1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d3de1fbcc820ef4eb0975a6090b007bc254960a3e5dba967df00c7b691431f6", "generation_seconds_batch": 23.784617349971086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0333168506622314, "target_z": -0.44921869628787636, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.704216718673706, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dd7010c16ea5997a0876dc4f7f35fde217a9f5e6b6fa14853c10b978b5bea8a1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "dd7e9d9543c19bee44dafc19a75f0dc2afcc3aeec3d087777562bae6e86ccdbe", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "832c5beb297cc8fffec98e25704772518eeb1a5b60f446216079e82b8cf4fb55", "generation_seconds_batch": 24.507337045972236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7707912921905518, "target_z": -0.13378902221590727, "ce": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.3514353930950165, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dd7e9d9543c19bee44dafc19a75f0dc2afcc3aeec3d087777562bae6e86ccdbe.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "dd8079f791031d0889bbd125bc1b700cb1870fc3d88fb1777cd5632151bf6422", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "816202dfc8a7690e1d3115edc888991ada074f8ff5eeead84a40adde5cd82295", "generation_seconds_batch": 21.054227849002928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.270936131477356, "target_z": 0.4667968839520206, "ce": 5.6413655281066895, "genuineness": 4.444586753845215, "identity": 0.6914229393005371, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/dd8079f791031d0889bbd125bc1b700cb1870fc3d88fb1777cd5632151bf6422.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "de79d618c5ff9da0a16fbc1a8a443c54967838ab3efed1ca89aeb054864af2d7", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "621306f12cbeec22eded355c90079636b8eb9644b1d7639c08da6244633a262f", "generation_seconds_batch": 23.479900148930028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4790059328079224, "target_z": 0.2167968839520206, "ce": 6.125868320465088, "genuineness": 4.645420074462891, "identity": 0.8074663877487183, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/de79d618c5ff9da0a16fbc1a8a443c54967838ab3efed1ca89aeb054864af2d7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "de7a03fe74aeb2afb7f8753c406d141cbb60bc55ef9f8b3727e31ad0435b99cb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7db51f4cec225d46294af194398a83baf050dca5b3b883a1f3e06a9e8b44cea9", "generation_seconds_batch": 23.289168171933852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2869019508361816, "target_z": -0.7539062320959588, "ce": 6.213237762451172, "genuineness": 4.268857479095459, "identity": 0.3988666534423828, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/de7a03fe74aeb2afb7f8753c406d141cbb60bc55ef9f8b3727e31ad0435b99cb.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "de7d0cdffba53eb8d9388c3f319807108a9e34ca552a794fafcc6f7280a24f77", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e1acccca783f737b98b610c2c9d6501a96686904533e1ea8a06698682490e4c3", "generation_seconds_batch": 17.06571057497058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.677322506904602, "target_z": -0.021484419760103037, "ce": 6.006728172302246, "genuineness": 3.986435890197754, "identity": 0.5989797115325928, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/de7d0cdffba53eb8d9388c3f319807108a9e34ca552a794fafcc6f7280a24f77.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "df0bc51d79c23c32c3497fb486e3b36d8d2613c949aa075a4da16fa6b0afd515", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e24b8616a99b4daa737ee6ef30022c490a6da96c739d0c27a2fcbd3017cb6fc", "generation_seconds_batch": 25.15223058196716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.381473183631897, "target_z": 0.333984419760103, "ce": 6.09409761428833, "genuineness": 4.68307638168335, "identity": 0.8357360363006592, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/df0bc51d79c23c32c3497fb486e3b36d8d2613c949aa075a4da16fa6b0afd515.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "dff86c15eba33b2c9fa1bfa5a5376297d432f4fa877e214387a95c6ab18dd00a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ece481fb73c9f9ba9fb66437d718b725f10015e8e50b0450a6a508d4870fa375", "generation_seconds_batch": 21.054227849002928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.673258662223816, "target_z": -0.016601629640154554, "ce": 6.117925643920898, "genuineness": 4.821148872375488, "identity": 0.6634330153465271, "hypothesis": "I've trouved the petite boîte. It was pretty fenêtre. We've regarded anything what.", "wer": 0.8421052631578947, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/dff86c15eba33b2c9fa1bfa5a5376297d432f4fa877e214387a95c6ab18dd00a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "e0432235812896edff235d24f63aa4fa441a87492b320f89c1ad4ba4be4bb648", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "4cc47058f2c2e4ec6f51aa857ba6eddd2dc762ae78ac9d7732646d68223dc8f8", "generation_seconds_batch": 18.520342689938843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7191802263259888, "target_z": -0.07177730122790212, "ce": 6.300607204437256, "genuineness": 4.262581825256348, "identity": 0.385145902633667, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0432235812896edff235d24f63aa4fa441a87492b320f89c1ad4ba4be4bb648.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e0637321f95f19054cce14fcff0e4982db7a13cd4d4ec2e27069f795e278743e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8f6a8504d995824554a41bc82552c919f1b1eaeafdb31ecd18d3ce05489799e8", "generation_seconds_batch": 15.976094212965108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7346229553222656, "target_z": -0.09033204691603607, "ce": 6.23706579208374, "genuineness": 4.595211505889893, "identity": 0.6962187886238098, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0637321f95f19054cce14fcff0e4982db7a13cd4d4ec2e27069f795e278743e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e0838a5b909ee13bd8f47771953fa8b003983b1a9cf774ff1d563d48286ddb70", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d95edcc0bcd06846ea4a7e1f2f48894f092caa2e5937ea3d2bfbc5c408d3003e", "generation_seconds_batch": 24.79242120392155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7439697980880737, "target_z": -0.10156246419191757, "ce": 6.062326908111572, "genuineness": 4.162164688110352, "identity": 0.30105501413345337, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0838a5b909ee13bd8f47771953fa8b003983b1a9cf774ff1d563d48286ddb70.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": "e08939bdd0461a0f6de503fbdf4e537bea1d025e6983e5c54969cf39075df117", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "f08ee92541337f05d418b1a1826d3d23b80d37424f50659d7cda91a1b3cdcf74", "generation_seconds_batch": 24.93912238604389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.878890037536621, "target_z": -0.26367181233585574, "ce": 6.213237762451172, "genuineness": 4.65797233581543, "identity": 0.6625311374664307, "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_TENS_reg_low/e08939bdd0461a0f6de503fbdf4e537bea1d025e6983e5c54969cf39075df117.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "e0c652528a3ff02b5dc03b673df64d45adcd996549ddf2744e7a5b8714505665", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9223ca1600769fe4ba8c1a460600b9a10b5c8b1af7629c0a4e25f876162f3ce5", "generation_seconds_batch": 25.230264256009832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7480336427688599, "target_z": -0.10644525431186605, "ce": 5.903473377227783, "genuineness": 4.055472373962402, "identity": 0.5499125123023987, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0c652528a3ff02b5dc03b673df64d45adcd996549ddf2744e7a5b8714505665.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "e0da30603d75d3a303bcaf3191cf19aaa5910f87b9625e7878420a112345ae91", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ac5703b3538fc4cef17c3840604bffd1890d0402f70d2ea9f87d498b76a516f1", "generation_seconds_batch": 16.91334024700336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5285850763320923, "target_z": 0.1572265580239897, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.7926371097564697, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0da30603d75d3a303bcaf3191cf19aaa5910f87b9625e7878420a112345ae91.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e0f34e96fdbd3f7cbb0c9f3086a8665567f14cbd5db087a5ae712d0866c9721b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ed16869cb2026da0b3663d1960817a882a6b8fe912498db326ec32b0455d54a", "generation_seconds_batch": 18.266433548997156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2237952947616577, "target_z": 0.5234375358080824, "ce": 5.982900142669678, "genuineness": 4.469690799713135, "identity": 0.6672686338424683, "hypothesis": "Y he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e0f34e96fdbd3f7cbb0c9f3086a8665567f14cbd5db087a5ae712d0866c9721b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e1298a2b96db3d7f8da8ddc1403dfe6c2f56e02133bf93e57f19f4d431dfe228", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d79adb5df4073e492b950c500c8171e08fb8504864ad18ebf374e235ad14cda9", "generation_seconds_batch": 22.441705646924675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9439119100570679, "target_z": -0.3417968839520206, "ce": 6.268836498260498, "genuineness": 4.017816066741943, "identity": 0.2121809422969818, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e1298a2b96db3d7f8da8ddc1403dfe6c2f56e02133bf93e57f19f4d431dfe228.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": "e12e610dec32afcdea1547a57a4f4ae36b0ca472bc53d0182b161a7e24c63b54", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "7b8b856df8f41db7a72b031c47242433ef0a6fc321c822c6eacd6ce9d01f3069", "generation_seconds_batch": 24.06995008594822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5163934230804443, "target_z": 0.17187507161616486, "ce": 6.181467056274414, "genuineness": 4.607763767242432, "identity": 0.6708333492279053, "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_TENS_reg_low/e12e610dec32afcdea1547a57a4f4ae36b0ca472bc53d0182b161a7e24c63b54.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "e1458e54e699f90fd7313e57fe8b872e71e78536670ea24424dadeab21466975", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9252d506eec022ef25d7417db2ca5675b1a89fc018b92be8cdabfdd11735965a", "generation_seconds_batch": 25.090283766039647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0755810737609863, "target_z": -0.5, "ce": 6.181467056274414, "genuineness": 4.262581825256348, "identity": 0.49233779311180115, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e1458e54e699f90fd7313e57fe8b872e71e78536670ea24424dadeab21466975.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e15cd2b735e60cf2d8a99212d4057f9b5d8f85cd2648c1eec2d499d0adbac21f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c814e4d0c7fd067f1d42518bd99da0ff80190bbb2f03e1d6c58ea3c467f83f33", "generation_seconds_batch": 23.074417750001885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6106750965118408, "target_z": 0.058593767904041215, "ce": 6.0146708488464355, "genuineness": 4.770940780639648, "identity": 0.7755695581436157, "hypothesis": "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_TENS_reg_low/e15cd2b735e60cf2d8a99212d4057f9b5d8f85cd2648c1eec2d499d0adbac21f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "e1a207b3bc87f7339bb44156012a8cd78ba0b4ad5c36387da411612b40289c11", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "72f8d17378cc7d2dfd308f9ca5821a56219f5cf8a5265ed09b5fda1e5374aba4", "generation_seconds_batch": 17.703527727047913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2966551780700684, "target_z": -0.7656249283838351, "ce": 6.062326908111572, "genuineness": 4.356721878051758, "identity": 0.5649926662445068, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e1a207b3bc87f7339bb44156012a8cd78ba0b4ad5c36387da411612b40289c11.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e1ba5263c081fc930d0a9d2662e9389fc286e9a57cc8043a684848abeef7cc87", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1b03f26764d513b2adad02edc882728219b745a65e56363d1599a3a2a16157b", "generation_seconds_batch": 24.2920363729354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5724748373031616, "target_z": 0.10449213826388666, "ce": 6.078212261199951, "genuineness": 4.595211505889893, "identity": 0.6563259363174438, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e1ba5263c081fc930d0a9d2662e9389fc286e9a57cc8043a684848abeef7cc87.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e27e0a866104a0a81894c6ecdea28171699f1a156a00f6ab49bd314e3824c275", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "5890267e9733962a9c33ad5f397546578254cb012a4a94336de56340cab90e4f", "generation_seconds_batch": 19.42186652997043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9292819499969482, "target_z": -0.32421869628787636, "ce": 6.252951145172119, "genuineness": 4.111956596374512, "identity": 0.389602929353714, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e27e0a866104a0a81894c6ecdea28171699f1a156a00f6ab49bd314e3824c275.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "e284064b9f9588c3bb08a5d46ca4eac87fd465d4dddd382cf197597ee7004a90", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfc4f9e479444be7068827e6f8d9805ef482d759575a2ed84f502ebc3ad432dd", "generation_seconds_batch": 24.71491583099123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3912265300750732, "target_z": 0.322265580239897, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.6946293115615845, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre, et 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_TENS_reg_low/e284064b9f9588c3bb08a5d46ca4eac87fd465d4dddd382cf197597ee7004a90.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "e352d25379884f8e6586d92ae945a428af0a0349076fec78540fe6ef1866964d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7704aa6b82806086c4029fe315e5286c295d9e03c2fe7524cec7d46697b7ec71", "generation_seconds_batch": 25.345918715931475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4920103549957275, "target_z": 0.20117181233585574, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.7775228023529053, "hypothesis": "I thought the petite boîte. Elle is pretty lay fenêtre. We regard on some what it contained.", "wer": 0.7894736842105263, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e352d25379884f8e6586d92ae945a428af0a0349076fec78540fe6ef1866964d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "e38959e925bb8ad5bc5ac9668105e97ab61586599aec06399d702cc49bcd5948", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "19334f162034d756d8f87bc521a9c1534ff8f6cabfd6b5175ea05cb729eda209", "generation_seconds_batch": 20.382125031086616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1471049785614014, "target_z": -0.5859373925757527, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.7473310232162476, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e38959e925bb8ad5bc5ac9668105e97ab61586599aec06399d702cc49bcd5948.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": "e39d85f9bd86d93f2a9ef133a4a5b9522dbb2907d091190aee6f5cdabbeb8482", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b91f3b5abb0f8b3802b6eb4b45f9208b3995023ed7e1fae2b5461bd091b1dbfb", "generation_seconds_batch": 26.802464232081547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5854791402816772, "target_z": 0.08886720988005152, "ce": 6.181467056274414, "genuineness": 4.68307638168335, "identity": 0.7159563302993774, "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_TENS_reg_low/e39d85f9bd86d93f2a9ef133a4a5b9522dbb2907d091190aee6f5cdabbeb8482.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e4d0acd84786e940b7a0d40e2db0d1e9e9e2d28042c8e35781ea535e56da70d1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "068b09c076f8af267137df8dd33aade527803f285ba87a4fffbe20e1eda00e05", "generation_seconds_batch": 24.248063872917555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7594125270843506, "target_z": -0.12011720988005152, "ce": 6.356205940246582, "genuineness": 4.2374773025512695, "identity": 0.34049326181411743, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e4d0acd84786e940b7a0d40e2db0d1e9e9e2d28042c8e35781ea535e56da70d1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "e4f495aeff0c127f7fc65cfac1b23de412f382094b70e2a9ea5d0eae4dcb79ba", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee13ca69a2a54b4daa5d6ae8aa020af5306c02be62274050ae87aac6e4f0337a", "generation_seconds_batch": 24.863072050036862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.043070077896118, "target_z": -0.4609373925757527, "ce": 6.0305562019348145, "genuineness": 4.262581825256348, "identity": 0.6855262517929077, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e4f495aeff0c127f7fc65cfac1b23de412f382094b70e2a9ea5d0eae4dcb79ba.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "e58003777aa9829f829ddd9b8517da173c8be1bb252695015dc867e1b8ce6b1a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30541aa93a8e9a13722c9c2f06e7f194e6b79c262f13ba3a48e0a091011e9fbc", "generation_seconds_batch": 24.132190837990493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9097754955291748, "target_z": -0.30078130371212364, "ce": 5.959072113037109, "genuineness": 4.162164688110352, "identity": 0.49587196111679077, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e58003777aa9829f829ddd9b8517da173c8be1bb252695015dc867e1b8ce6b1a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "e60b477690d176a808e6fe98735b7c8b947bdc46e1ad74b8e3862d41309a6658", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa4887ae17b841ce6e7aefee6ac221a79c8229e6783e01540475b791575873ce", "generation_seconds_batch": 26.772241360973567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4806314706802368, "target_z": 0.21484376790404122, "ce": 5.911416053771973, "genuineness": 4.319066047668457, "identity": 0.2698978781700134, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e60b477690d176a808e6fe98735b7c8b947bdc46e1ad74b8e3862d41309a6658.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": "e64152ebc088b76a984827a856da0139c9cdaa650eaecf13f752b0d7e3ec184f", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "f3896ccef064e6350de6f7cf292b7857662488335565c984f9bf37ee1d48cc5b", "generation_seconds_batch": 21.78954419097863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4660015106201172, "target_z": 0.23242195556818546, "ce": 5.927301406860352, "genuineness": 4.6956281661987305, "identity": 0.7575252056121826, "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": 0.120000000000001, "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_TENS_reg_low/e64152ebc088b76a984827a856da0139c9cdaa650eaecf13f752b0d7e3ec184f.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very relaxed and unstrained. 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": "13117623bd4d9d336f20a360849fa47b461ec1f4d490f954ede8d19eabcb0a1b", "generation_seconds_batch": 18.755763795110397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7183674573898315, "target_z": -0.07080074320391241, "ce": 6.356205940246582, "genuineness": 4.6203155517578125, "identity": 0.39087480306625366, "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_TENS_reg_low/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e71eb5c58ec53632e6286a2990b64819e5ddfd4a6fe2a741d737c7e738fc7044", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "049168cb959a616ea43d0e2e589b150e223b87680576a96a34b84f598d155c39", "generation_seconds_batch": 20.040056150988676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9682950973510742, "target_z": -0.3710937679040412, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.12652766704559326, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e71eb5c58ec53632e6286a2990b64819e5ddfd4a6fe2a741d737c7e738fc7044.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "e74e836130e3ff2c84a3a71110974333c5db03bbdcce73b248d60e846065795e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cebd37754b5a32c2b9ddad618a17d01c6d2ce236d105175b7584a74ead290da5", "generation_seconds_batch": 24.75148405705113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7902978658676147, "target_z": -0.1572265580239897, "ce": 6.197352409362793, "genuineness": 4.2751336097717285, "identity": 0.3838900029659271, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e74e836130e3ff2c84a3a71110974333c5db03bbdcce73b248d60e846065795e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "e7a75ddc2d9ec55eda01e0968d8c78e99aca3091600bead91eed9bca33373e05", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6abce2f240c405ff8601a80bac8bb86c507cf3f0ce657d25b7a87b198501e57", "generation_seconds_batch": 17.45856888103299, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": -0.644531160479794, "ce": 6.038498878479004, "genuineness": 4.582659721374512, "identity": 0.5158517956733704, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e7a75ddc2d9ec55eda01e0968d8c78e99aca3091600bead91eed9bca33373e05.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "e7ba1bd0ff39fac9082e86adce6cb2bbb2daa19ad417469220e7c6546f636936", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba2531109d8419de806249ac8531b1f0e604f03aec1e287cd803b72182e913e2", "generation_seconds_batch": 26.127805741969496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083998441696167, "target_z": 0.6914062320959588, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.7937761545181274, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e7ba1bd0ff39fac9082e86adce6cb2bbb2daa19ad417469220e7c6546f636936.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": "e88b58c32f452344d247cf0bf6bff2b6836ec3e0d0758b0e82a7914b05e9c468", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dae8b2ba91def716bd5afe23dd8aecc2ef5a49ff1db5b37e9b1bc046f15a934c", "generation_seconds_batch": 19.06373547494877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9650440216064453, "target_z": -0.36718753580808244, "ce": 6.300607204437256, "genuineness": 4.507347106933594, "identity": 0.5336896181106567, "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_TENS_reg_low/e88b58c32f452344d247cf0bf6bff2b6836ec3e0d0758b0e82a7914b05e9c468.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "e8a236de17ccd7fc0a64c78796ae2df431ee2bca453797e7e6026ee75981a418", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de92179a83524922409996ca0b9cb2f6cf301aa4d10b42cc098b2be836b58716", "generation_seconds_batch": 26.98647429398261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4253629446029663, "target_z": 0.28125, "ce": 5.561938762664795, "genuineness": 4.494794845581055, "identity": 0.6854379773139954, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e8a236de17ccd7fc0a64c78796ae2df431ee2bca453797e7e6026ee75981a418.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "e8ba1f9445525391544b1b0f48dd4e122dfe56517ee7de08caeee82bfbcd71f3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4437e1d947b7e6b2e5481b9a7c3000d269b56b3a871ff4a3680d05239233444a", "generation_seconds_batch": 27.164548998931423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4895720481872559, "target_z": 0.20410148640782483, "ce": 5.5539960861206055, "genuineness": 4.093128204345703, "identity": 0.5448080897331238, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e8ba1f9445525391544b1b0f48dd4e122dfe56517ee7de08caeee82bfbcd71f3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e8ed783c2a9824b8ed0e4f8f8063ac4c9f60189c44b6c7fab95fb90365029606", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "90bc80cbe9485f5c0f80fd9362d20f90f240f4de7b35d435c0fbecfd251911b4", "generation_seconds_batch": 17.42860073398333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8918944597244263, "target_z": -0.2792968839520206, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.48416590690612793, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e8ed783c2a9824b8ed0e4f8f8063ac4c9f60189c44b6c7fab95fb90365029606.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "e91773d80dba6eeedbaf72e41380d1fac227207501ed076135ef5066f313122e", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "46e7d107538f8f7135df4467b35254b8b34127d6380b057402cb068eabecfc0e", "generation_seconds_batch": 24.442128799972124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4237374067306519, "target_z": 0.2832031160479794, "ce": 6.038498878479004, "genuineness": 4.306513786315918, "identity": 0.7741321325302124, "hypothesis": "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_TENS_reg_low/e91773d80dba6eeedbaf72e41380d1fac227207501ed076135ef5066f313122e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "e9258ae5ae97657d282dcc0ac94524a8f8d1a537783d33c3b8ebfff9ac22f1c4", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "a61e89f3d0b81d91b12fbdea5ba37ffd8021bbf9d0d521033fb2bc41b4c02c72", "generation_seconds_batch": 15.904755450901575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8000510931015015, "target_z": -0.16894525431186605, "ce": 6.332377910614014, "genuineness": 4.093128204345703, "identity": 0.25202804803848267, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e9258ae5ae97657d282dcc0ac94524a8f8d1a537783d33c3b8ebfff9ac22f1c4.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": "e97bfd12861c5a0b3bf317eefb4f233e522c96c19971b23d233432574d56df4b", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "124502bd03e65486332f8f7da15e01e5c9ec3d0cf4eeb1e1c2af494c78068da6", "generation_seconds_batch": 26.570803523994982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6200220584869385, "target_z": 0.04736320739582999, "ce": 6.149696350097656, "genuineness": 4.532451152801514, "identity": 0.7321231961250305, "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_TENS_reg_low/e97bfd12861c5a0b3bf317eefb4f233e522c96c19971b23d233432574d56df4b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "e9911829b5e51aed1f84d35680df605ccb2613692e6748eec76109cf6b064c9c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bfe11d6714225e0e86af6e6ca27ff26a98d8c1c9e934f55384b787bc80c265e5", "generation_seconds_batch": 22.82235076301731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6594414710998535, "target_z": -0.0, "ce": 6.38003396987915, "genuineness": 4.595211505889893, "identity": 0.7087697982788086, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e9911829b5e51aed1f84d35680df605ccb2613692e6748eec76109cf6b064c9c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "e9c74ae4ff1fec90f2390bf1267f935d0fd795dcd5f2299d505f37e8fa07dfa6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86910542e55417c47a835d6c9102b3bd45935331499e75e1d7d26c5f187b730c", "generation_seconds_batch": 24.595487658050843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7334038019180298, "target_z": -0.08886720988005152, "ce": 6.3641486167907715, "genuineness": 4.042920112609863, "identity": 0.5343989133834839, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/e9c74ae4ff1fec90f2390bf1267f935d0fd795dcd5f2299d505f37e8fa07dfa6.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very relaxed and unstrained. 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": "9c61b7ff0e79509d32d98f0246bff0d42ea78043833b21e16eff44be3c5f61b9", "generation_seconds_batch": 17.11650668596849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8236215114593506, "target_z": -0.19726558023989696, "ce": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.5623934268951416, "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_TENS_reg_low/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.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": "ea61526bebc02c1c34ccb6a82357d397fec1847be3f1ce660cbf110e431d6bba", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c954761d50c24b6a2d04a32874b3ad3317b0a0713f445f717c7e68606d6d7dd", "generation_seconds_batch": 18.839654837967828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.687888503074646, "target_z": -0.03417967407196909, "ce": 6.022613525390625, "genuineness": 4.519899368286133, "identity": 0.764180064201355, "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_TENS_reg_low/ea61526bebc02c1c34ccb6a82357d397fec1847be3f1ce660cbf110e431d6bba.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "ea61b4b6fdaeba85c2ba15537ba50e562f8c4e166c752f2a2b5db1202808fa09", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "399df5dac9263252261a15528de007302c43583ba55605a15336a7d027efb1ad", "generation_seconds_batch": 18.620863603893667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4383673667907715, "target_z": 0.26562492838383517, "ce": 5.649308204650879, "genuineness": 4.545003414154053, "identity": 0.7794493436813354, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ea61b4b6fdaeba85c2ba15537ba50e562f8c4e166c752f2a2b5db1202808fa09.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "ea779ccd4c9f5a7e89b5f4b0a7002dcf0cfefdafdbb026c703360cc0567a5acc", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a52d07085b7d5c0ad26769fa00b930193f738e6cf2fdce4a68827dc913e79113", "generation_seconds_batch": 18.32406831299886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5781642198562622, "target_z": 0.09765623209595879, "ce": 6.284721851348877, "genuineness": 4.419482707977295, "identity": 0.6003454327583313, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ea779ccd4c9f5a7e89b5f4b0a7002dcf0cfefdafdbb026c703360cc0567a5acc.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "ea9a5e73b872b3a71a00e10fd6bdbc65bfd13fd754ada989e14ba6edb1b3f106", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "97fba36eb932d6dea9b3d441af9b071f9c90ea573209ce7dc7671b40ad7318a3", "generation_seconds_batch": 23.229884275933728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191284418106079, "target_z": 0.5625, "ce": 6.252951145172119, "genuineness": 4.519899368286133, "identity": 0.627548098564148, "hypothesis": "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_TENS_reg_low/ea9a5e73b872b3a71a00e10fd6bdbc65bfd13fd754ada989e14ba6edb1b3f106.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "eb3ed5404781096b0687a10bd081c1bde7738e321c7c6f7e29c3b5b37a621224", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14de33617c19027c634273220da675449d398b80834641a75fb0f52cb90c3f5a", "generation_seconds_batch": 23.31847012799699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.531023383140564, "target_z": 0.1542968839520206, "ce": 5.903473377227783, "genuineness": 4.20609712600708, "identity": 0.590447187423706, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/eb3ed5404781096b0687a10bd081c1bde7738e321c7c6f7e29c3b5b37a621224.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "eb5481caddd07e19da6c620bd01ffd0642ab1c01b16035c7f60648810c8edeb1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e0c660aa385a22bac55cbbc8ca8ff54bfa1a2470f1d95d4ebaf90b34cfd3764f", "generation_seconds_batch": 23.08492805901915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6214443445205688, "target_z": 0.04565430247001288, "ce": 6.324435234069824, "genuineness": 4.1998209953308105, "identity": 0.6457487344741821, "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_TENS_reg_low/eb5481caddd07e19da6c620bd01ffd0642ab1c01b16035c7f60648810c8edeb1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "eb8928efcfa8bede77d836ef74cd7ed8295f5bdc47d93e55e8ea87a358f643b6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fef04f5461e84b2c6a2b1ed33be2d8cad3ace66341ed61f5a85300d5e59cdaa", "generation_seconds_batch": 17.0072726609651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.397728681564331, "target_z": 0.3144531160479794, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.6532776355743408, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/eb8928efcfa8bede77d836ef74cd7ed8295f5bdc47d93e55e8ea87a358f643b6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "ebaa9f7ff559868918208086e04324ab1833df1812fe304af3996cc2c1577888", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6954ee7fa214a2442333170dea8625ea5096745e421cb646856205773cf91097", "generation_seconds_batch": 25.18278258200735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": -0.644531160479794, "ce": 6.078212261199951, "genuineness": 4.250029563903809, "identity": 0.48056232929229736, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ebaa9f7ff559868918208086e04324ab1833df1812fe304af3996cc2c1577888.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "ebc80c1a350d3692358e9717676612b20df81d4e76d6c92b2bf040d6c4dea122", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3e371b4004c025add48430cc3520a905ca5674612418f347cb8763187e084561", "generation_seconds_batch": 23.84099049901124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.484695315361023, "target_z": 0.20996097778409273, "ce": 6.078212261199951, "genuineness": 4.381826400756836, "identity": 0.8015414476394653, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/ebc80c1a350d3692358e9717676612b20df81d4e76d6c92b2bf040d6c4dea122.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very relaxed and unstrained. 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": "e98cd0daad933970fb30429200f480c8280920fca237b45d944834a044fb72ef", "generation_seconds_batch": 18.15688769193366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5623151063919067, "target_z": 0.11669925679608759, "ce": 6.284721851348877, "genuineness": 4.3881025314331055, "identity": 0.8302332758903503, "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_TENS_reg_low/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "ebf877209c6dfafc4e0e165ee8898a551c859511ce1632ff9f944c1ae531735d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "945d55e21c8ef1de102bd0f188b030ee6af4b81c4485932eee8cdf419215b521", "generation_seconds_batch": 16.565960846026428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.53508722782135, "target_z": 0.14941409383207213, "ce": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.6723145246505737, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ebf877209c6dfafc4e0e165ee8898a551c859511ce1632ff9f944c1ae531735d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "ec2d346f292e99f595906d981a94260e568494f44a3bc20460d88b59822ea6a5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a034b5183232e0fd6e5525fd2a01922d64bb5336cc73404c352bd26c398f733", "generation_seconds_batch": 26.549365081940778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2920682430267334, "target_z": 0.4414062320959588, "ce": 5.8875885009765625, "genuineness": 4.482243061065674, "identity": 0.5899507999420166, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ec2d346f292e99f595906d981a94260e568494f44a3bc20460d88b59822ea6a5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "ec5509053633c9686add6364f37d3c35da65bd93218a0b27b5673fbb71fd93ae", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae99341e058691ba87ec208a72ce05e3fc2e0aee9e0bd97ede9a99fc2fa51b31", "generation_seconds_batch": 22.578386905952357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.355175018310547, "target_z": -0.8359376790404122, "ce": 5.974957466125488, "genuineness": 4.218649387359619, "identity": 0.6388916373252869, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ec5509053633c9686add6364f37d3c35da65bd93218a0b27b5673fbb71fd93ae.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "ec805cb11433999334a0888fa39b1472bf663947f81366bc8fec557b4149f54f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "346fe46e1884849303078ec55f5928233c5db9a619fd910ad348e08f3df849d4", "generation_seconds_batch": 17.09776693896856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8154938220977783, "target_z": -0.1875, "ce": 5.911416053771973, "genuineness": 4.061748504638672, "identity": 0.5120893716812134, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ec805cb11433999334a0888fa39b1472bf663947f81366bc8fec557b4149f54f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "eca8ae335d5d3fa03028022ff81f2612d75cbc7552273faa0821cadddfd20426", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2a42e917eb11b4cd40c533e147b6724ce582ff019039b6ae46fa4c236c741a73", "generation_seconds_batch": 18.182384665007703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3733454942703247, "target_z": 0.34375, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.5778332352638245, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/eca8ae335d5d3fa03028022ff81f2612d75cbc7552273faa0821cadddfd20426.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "ecd9875af559bf554b83826da5aa75a2780a4367b9521468c2f7e17649345361", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c350bdd3cc968d1a8a3896d35f98e4a082066232f0e227847539da0e993b9120", "generation_seconds_batch": 23.77860659500584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6318072080612183, "target_z": 0.033203116047979396, "ce": 5.974957466125488, "genuineness": 4.582659721374512, "identity": 0.6067935228347778, "hypothesis": "Die Tür 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_TENS_reg_low/ecd9875af559bf554b83826da5aa75a2780a4367b9521468c2f7e17649345361.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "ecfcd49414a329f554b0d5d8eebb2f66b6021697f4d69bb509931cde0c6da27f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39ce152beb52bb7a3487e4ef4c28eaccb83652165cc36da2ce3f706d92a11d29", "generation_seconds_batch": 23.620418836013414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0186870098114014, "target_z": -0.4316406518560618, "ce": 6.284721851348877, "genuineness": 4.005263805389404, "identity": 0.4203585982322693, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ecfcd49414a329f554b0d5d8eebb2f66b6021697f4d69bb509931cde0c6da27f.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": "ed202b7008e64c31037934541da61b1a3d9bf65bc37918da3daab9f4eabf2d1f", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The keys are on the kitchen table.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9736a78f65d0161d08a04a2a6d2008388294d41100f52c368280350af50515c4", "generation_seconds_batch": 20.51433985296171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7033312320709229, "target_z": -0.052734419760103034, "ce": 6.38003396987915, "genuineness": 4.074300289154053, "identity": 0.31448134779930115, "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_TENS_reg_low/ed202b7008e64c31037934541da61b1a3d9bf65bc37918da3daab9f4eabf2d1f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "ed2cb9aa13f920d9b414dfff65d2d25c625e0562c0c5a71f4895dc7c97f79e0a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b499c52991a935d6170d5a87a3ade4ba8cdfa9c3447f0d1578297c79e046166", "generation_seconds_batch": 22.331478249048814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2251312732696533, "target_z": -0.6796875358080824, "ce": 6.181467056274414, "genuineness": 4.570107460021973, "identity": 0.1414089798927307, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ed2cb9aa13f920d9b414dfff65d2d25c625e0562c0c5a71f4895dc7c97f79e0a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "ed8a23f3ea1e6506164c10a211ba0f3e182bfffca8ec2771d882c487d8de571b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0cd58893686bf8c2ed9060f5c75031dae0150ed2d2c3a27b18148a1ba48ba13", "generation_seconds_batch": 24.38807904999703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7195866107940674, "target_z": -0.07226558023989696, "ce": 6.268836498260498, "genuineness": 4.306513786315918, "identity": 0.33615249395370483, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ed8a23f3ea1e6506164c10a211ba0f3e182bfffca8ec2771d882c487d8de571b.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": "edbdc02f2caeb93ac5e243088ab2994d57233e2326956b35625d26c177b8f829", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f83826650950c0a39e886c2f1abea196560980872c1dff6c23aa1b4c70340bca", "generation_seconds_batch": 21.362750987056643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3197025060653687, "target_z": 0.4082031160479794, "ce": 5.879645347595215, "genuineness": 4.6705241203308105, "identity": 0.847256064414978, "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_TENS_reg_low/edbdc02f2caeb93ac5e243088ab2994d57233e2326956b35625d26c177b8f829.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "ede907ed2f7f78ca5f6111fe441f760450c99b0f3fa3319132d7c422383b5519", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a7e931c778c069b2720352e5ba7337351fd949a501f123acb45ef3239207523", "generation_seconds_batch": 23.43836496397853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5570321083068848, "target_z": 0.12304688395202061, "ce": 5.959072113037109, "genuineness": 4.507347106933594, "identity": 0.4237760305404663, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ede907ed2f7f78ca5f6111fe441f760450c99b0f3fa3319132d7c422383b5519.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "ee5b3257790c8e605413b1b4d206999522d58e79145fc4fa50846af204aa5919", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "603985b33ca0ecab33ddedd15936b08b0fe91b44bc18a2d147a2a1dbeac612b9", "generation_seconds_batch": 18.20107739896048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7520976066589355, "target_z": -0.11132818766414425, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.6737179160118103, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ee5b3257790c8e605413b1b4d206999522d58e79145fc4fa50846af204aa5919.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "ee8aac650b7c14c1bb04a703f7605d6967ee367c1462d18c84bcda9af570812a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33744e86b5f8b3c77f54a124405b489a1eb0c3fcb4f1abedac002a53798cffad", "generation_seconds_batch": 26.687958552036434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.40097975730896, "target_z": 0.3105468839520206, "ce": 5.720792293548584, "genuineness": 4.394378662109375, "identity": 0.5143897533416748, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ee8aac650b7c14c1bb04a703f7605d6967ee367c1462d18c84bcda9af570812a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "eeb77cdeb0fcad5d101ec8e2ad1f562706dd16c7fe9d0df0549a831a59aa1bef", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a8911f5738fcdd94705ef7904491522332d79480509b5f6875d1aca94f37c0c", "generation_seconds_batch": 23.282547803013586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4725037813186646, "target_z": 0.22460934814393818, "ce": 5.863759994506836, "genuineness": 4.042920112609863, "identity": 0.5497339963912964, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/eeb77cdeb0fcad5d101ec8e2ad1f562706dd16c7fe9d0df0549a831a59aa1bef.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "ef6bf1e0be98a2de8af85d56202c6d419e8a41f86283d716731aa0403c385d41", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7de793e86820ca3a4d5ba7cdc805b2c827c835613b1493052753892a4ae4ed7a", "generation_seconds_batch": 23.70311780902557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6921555995941162, "target_z": -0.03930667531407985, "ce": 6.284721851348877, "genuineness": 4.494794845581055, "identity": 0.25327005982398987, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/ef6bf1e0be98a2de8af85d56202c6d419e8a41f86283d716731aa0403c385d41.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "efcf5b55b95f8d2f525ec75783be383bcde2e68e9628d30a2572ff763ed62a0d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "410603c386e24eb4e9aced7d52f88d487eeb74b83ae2e07c18f9f462936d5453", "generation_seconds_batch": 18.502930148970336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5033891201019287, "target_z": 0.1875, "ce": 6.197352409362793, "genuineness": 4.570107460021973, "identity": 0.677657961845398, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/efcf5b55b95f8d2f525ec75783be383bcde2e68e9628d30a2572ff763ed62a0d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "f004176f9e9fc36cf60aaa2e5e1aea2cd2e173c699a4224fea657ef029ed0921", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2940c16857cd79232d0e509965cb2f892775c6f54146f5edbb867d9a4945dfd2", "generation_seconds_batch": 17.28567596094217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.45920991897583, "target_z": -0.9609376790404122, "ce": 5.990842819213867, "genuineness": 4.18099308013916, "identity": 0.3445294201374054, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f004176f9e9fc36cf60aaa2e5e1aea2cd2e173c699a4224fea657ef029ed0921.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "f01646e29b740a79b76b3928cd5bfa7fcf9f00f661f8f28e91068efe540c27b9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d1b07861612d60d55f1aee5c4f05a8ad71485bb38a00b8e92a4a94c0467e1380", "generation_seconds_batch": 25.566503594978712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4497461318969727, "target_z": 0.2519531160479794, "ce": 6.332377910614014, "genuineness": 4.381826400756836, "identity": 0.5637202262878418, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f01646e29b740a79b76b3928cd5bfa7fcf9f00f661f8f28e91068efe540c27b9.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": "f053cd63d28c73a33f2151ea1cd1fe19e25c0bef8f902357c4496a9a10f7aab0", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4f9ce38c46c5633107b486107d7b551b4813dce0bacb5162caf4d985299f87e", "generation_seconds_batch": 22.939864960033447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5976707935333252, "target_z": 0.07421869628787636, "ce": 5.998785495758057, "genuineness": 4.607763767242432, "identity": 0.7966277599334717, "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_TENS_reg_low/f053cd63d28c73a33f2151ea1cd1fe19e25c0bef8f902357c4496a9a10f7aab0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "f0e52b110e75ff2bdca96bb11f2524b0b6b9cf5f7960d73a55b16a39f411e510", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c659e0ab4ce8e080e98e80f7954fc186679daa9332a03fa19228476d02f571d", "generation_seconds_batch": 24.36815634998493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4269884824752808, "target_z": 0.2792968839520206, "ce": 6.419747352600098, "genuineness": 4.6956281661987305, "identity": 0.4790270924568176, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f0e52b110e75ff2bdca96bb11f2524b0b6b9cf5f7960d73a55b16a39f411e510.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "f12a19c5c817a80cdbe61a02061b0f8b7a0c4d3de806cfb215d50bafca84e3b9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf7c48eb35e3f88511b959d9b998f966140c1e5d701c044e3023e82897e7459a", "generation_seconds_batch": 25.566488189040683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8057405948638916, "target_z": -0.17578130371212364, "ce": 6.197352409362793, "genuineness": 4.425758361816406, "identity": 0.5970262289047241, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f12a19c5c817a80cdbe61a02061b0f8b7a0c4d3de806cfb215d50bafca84e3b9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "f163f494d7269fa76ee27f640a8345f68806919fcc853dcc3e952228125c705b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f0c06212ebc5907b9099e75005645397d5ee83636a8e91c981549f28d672843", "generation_seconds_batch": 25.052163051092066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.917903184890747, "target_z": -0.3105468839520206, "ce": 6.149696350097656, "genuineness": 4.532451152801514, "identity": 0.5170631408691406, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f163f494d7269fa76ee27f640a8345f68806919fcc853dcc3e952228125c705b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f185a2ddd83d1da7a0bf931fa196a514222c04d0129eb327d5977d4bb815ea1a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed08b6fc80d1f9231355d827d37f02de926e3b02c4b313b05724829742c8a4ce", "generation_seconds_batch": 27.267675091978163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3148258924484253, "target_z": 0.41406246419191756, "ce": 5.649308204650879, "genuineness": 4.595211505889893, "identity": 0.7725406885147095, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f185a2ddd83d1da7a0bf931fa196a514222c04d0129eb327d5977d4bb815ea1a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "f19dd15ef5a9c15c1d395185aeede1bc94e35667a2670c99083e30eb0e894a95", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a44d91b05bd08adb7b04bf5b1363c504874e86fefbbe672639f8cf0649175e6f", "generation_seconds_batch": 16.246735175023787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4221118688583374, "target_z": 0.2851562320959588, "ce": 6.038498878479004, "genuineness": 4.143336772918701, "identity": 0.6522964239120483, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f19dd15ef5a9c15c1d395185aeede1bc94e35667a2670c99083e30eb0e894a95.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "f1a093635562077cd3eb4bf668575a21002a4586d5d37f659f7042565bb489d5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d73a84e7e2c3b9acfe6d1393f26cabe6c15ea5d887efc5947fcea49370abe1e", "generation_seconds_batch": 18.45995614095591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4432439804077148, "target_z": 0.259765580239897, "ce": 5.998785495758057, "genuineness": 4.287685871124268, "identity": 0.3558458089828491, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f1a093635562077cd3eb4bf668575a21002a4586d5d37f659f7042565bb489d5.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f1fec133b08056b98b6abc00d1b75488fb902de6a0d296ca26f3641fa00ce14b", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b05c5214d47f99c10169ab906b90f3f207cec65a5a8bf63e40f923052c3c3d40", "generation_seconds_batch": 19.314193704980426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.57003653049469, "target_z": 0.10742181233585575, "ce": 5.959072113037109, "genuineness": 4.632867813110352, "identity": 0.7985730171203613, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f1fec133b08056b98b6abc00d1b75488fb902de6a0d296ca26f3641fa00ce14b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "f21a53a13b32a7917a3ee60a122197088e5655b9ea92f25c973a40fb56ca354d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b3ecf85b6b5abead388398a61988d3df1d95fde758012589e503e15af5abf77", "generation_seconds_batch": 25.347713648923673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6399348974227905, "target_z": 0.02343753580808243, "ce": 5.911416053771973, "genuineness": 3.948779582977295, "identity": 0.590976893901825, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f21a53a13b32a7917a3ee60a122197088e5655b9ea92f25c973a40fb56ca354d.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very relaxed and unstrained. 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": "1bc9612d54bb24ef47394aa32ba23cfed26b6fe1afdb55823105c39b55aecb05", "generation_seconds_batch": 17.960630608955398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6439987421035767, "target_z": 0.018554745688133947, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.79905104637146, "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_TENS_reg_low/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "f2328a60eae61641c231e1a2c8189647d1a9108af8cb0ceab57a8cf8ec4b1bd9", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c524baa6cf094122f293b1dac4d7318863b52305e50226c96ad4e935b4eb1ce", "generation_seconds_batch": 22.899425540002994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6106750965118408, "target_z": 0.058593767904041215, "ce": 5.974957466125488, "genuineness": 4.595211505889893, "identity": 0.7015894651412964, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f2328a60eae61641c231e1a2c8189647d1a9108af8cb0ceab57a8cf8ec4b1bd9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "f2a8a11774e4522a648ce1be9dab3fc22be0e8c5d193d983770d61c2cfc5935f", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "c3468de3caf3fcfb81a44adb3b07419c297ac7e009b7cd1735f643fa79ca3d07", "generation_seconds_batch": 17.887065967079252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.726901650428772, "target_z": -0.08105474568813395, "ce": 6.372091293334961, "genuineness": 4.068024158477783, "identity": 0.47419142723083496, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f2a8a11774e4522a648ce1be9dab3fc22be0e8c5d193d983770d61c2cfc5935f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "f2a90308bc0f4b620914eb51b554db25a4ab099691858fda8c0844f58a6eba6a", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fef24e93db5e4c41b8b5f5b004f15df237436e5fade95ce42f9bc43912279801", "generation_seconds_batch": 18.899313492001966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7334038019180298, "target_z": -0.08886720988005152, "ce": 5.935244083404541, "genuineness": 4.099404335021973, "identity": 0.4807376265525818, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f2a90308bc0f4b620914eb51b554db25a4ab099691858fda8c0844f58a6eba6a.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "f2db6f7a119371f8ab4699e260f16289bbc795eacd46807eb52c2c195efbbfb6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59980484efed25f749c80273e51bd5a2e24f2acecde7e2164086ce292e34da74", "generation_seconds_batch": 25.775045663001947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2823150157928467, "target_z": 0.45312492838383517, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.47923603653907776, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f2db6f7a119371f8ab4699e260f16289bbc795eacd46807eb52c2c195efbbfb6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "f2ef3330603350984434a99e427cee4bc97781b4427fb8d67bc2f7a57b775246", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "e5ceb655e5ba3295cd1a5ff6e22e0bd834228dfa9e39845b33de545f92758463", "generation_seconds_batch": 20.05072914902121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4757548570632935, "target_z": 0.2207031160479794, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.5475941896438599, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f2ef3330603350984434a99e427cee4bc97781b4427fb8d67bc2f7a57b775246.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "f3118538a161e2539310422f7626b0b25d3e3ed32f1c73717f7b7d53c30bc4d0", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "dfe2803e3c6b479c53895792d9abe2471f2fa0b76ffbca7c337773b827156f85", "generation_seconds_batch": 15.904755450901575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6817927360534668, "target_z": -0.026855488892046368, "ce": 6.141753673553467, "genuineness": 3.9550557136535645, "identity": 0.3247588872909546, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f3118538a161e2539310422f7626b0b25d3e3ed32f1c73717f7b7d53c30bc4d0.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": "f31c99b445af146e2df95895b75f702124ba80c58e0b7c1a1beb7c2fd1c585b9", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Your message arrived after I had left.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "577ef72763d23c0f9659fb93162e11abeeb6b9e120bdb3e67d0a695f6d6a2388", "generation_seconds_batch": 27.31595028808806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8593835830688477, "target_z": -0.24023441976010304, "ce": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.38273316621780396, "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_TENS_reg_low/f31c99b445af146e2df95895b75f702124ba80c58e0b7c1a1beb7c2fd1c585b9.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "f3860565f5afc68617a6dde970831befe53590a2423babbd460212f2793553e6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d684dc8bf7c0e11bac7882b573d3707397b4cb7be1d951910c90ae21a0b2ce2c", "generation_seconds_batch": 22.897065024939366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.327830195426941, "target_z": 0.39843753580808244, "ce": 6.3641486167907715, "genuineness": 4.413206577301025, "identity": 0.41768091917037964, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f3860565f5afc68617a6dde970831befe53590a2423babbd460212f2793553e6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "f399ec4c5193e5db467db2d9ab6ce2ade29690d04d1aeb5d96a663706376c376", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "db33024b0297fa667f79a6a27453a051ea0f2439b9ad4f85c76f62c0c1b679ff", "generation_seconds_batch": 18.440039876033552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.355175018310547, "target_z": -0.8359376790404122, "ce": 6.0146708488464355, "genuineness": 4.231201171875, "identity": 0.6208152770996094, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f399ec4c5193e5db467db2d9ab6ce2ade29690d04d1aeb5d96a663706376c376.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "f3ccae25f06d8aa13ab9033087275958f74cdfe9123b04be18ccb75a6c08d51f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b910b420cb7303963bb06227bacc6ce238a8d0e02ed1d99055526cedd610d17b", "generation_seconds_batch": 18.369528984068893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9032732248306274, "target_z": -0.29296869628787636, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.69179368019104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f3ccae25f06d8aa13ab9033087275958f74cdfe9123b04be18ccb75a6c08d51f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "f4043bcd6ce787941d41db7a4a290bca944fa342520f49fcaa42b72ba2720ad1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dc82ec0518d50e3c30c25e9cfc0eafb1237665e428d1f7ea7a3b42bc4fe937b", "generation_seconds_batch": 18.899313492001966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.680573582649231, "target_z": -0.025390651856061822, "ce": 5.95112943649292, "genuineness": 4.011539936065674, "identity": 0.4960065484046936, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f4043bcd6ce787941d41db7a4a290bca944fa342520f49fcaa42b72ba2720ad1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "f4414541537b3be6884083d81553d6da37cf14105e978458f68647ec6572d24e", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "112a1e3caf0498e91dc98c047afbc133b25ffa385957a0ed8a222391dd7d6008", "generation_seconds_batch": 19.49372464697808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7707912921905518, "target_z": -0.13378902221590727, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.4712664484977722, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f4414541537b3be6884083d81553d6da37cf14105e978458f68647ec6572d24e.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f44829a0dd382855c2c2fe91c2abc5cd6206d36e6cf28dbf2e9c67f927cb1dda", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "327242a9f088f8b70e67c924d23e31ddbf4ddcc7cfce5612b7b1f9222b08c1f9", "generation_seconds_batch": 25.03373357397504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3489623069763184, "target_z": 0.3730468839520206, "ce": 5.824047088623047, "genuineness": 4.645420074462891, "identity": 0.769034206867218, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f44829a0dd382855c2c2fe91c2abc5cd6206d36e6cf28dbf2e9c67f927cb1dda.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "f46e7e8666f9c9db1a89ca3fdd1fe950cf9114dc650e6856adbfabe7b609b8c8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2199655e58fc5f09d8b3757f6e3775da9269e1a25297000dc9d5d149033d3280", "generation_seconds_batch": 24.87286729400512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9422863721847534, "target_z": -0.3398437679040412, "ce": 6.038498878479004, "genuineness": 4.3127899169921875, "identity": 0.6891411542892456, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f46e7e8666f9c9db1a89ca3fdd1fe950cf9114dc650e6856adbfabe7b609b8c8.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": "f59545d4ff56a8d5a59b964e4cd5fba25ea2dc0217f718fc754cf60a7cddbc3b", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9e8199e8e7afcf0e537dc6e0b3c75970bcaf43964dd38a56e83f63b8100a6f96", "generation_seconds_batch": 21.362750987056643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3700944185256958, "target_z": 0.3476562320959588, "ce": 5.935244083404541, "genuineness": 4.808597087860107, "identity": 0.6537337303161621, "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_TENS_reg_low/f59545d4ff56a8d5a59b964e4cd5fba25ea2dc0217f718fc754cf60a7cddbc3b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "f5ba6234103650e725017915809f167c10de698ede7b2e2e687600c2bcbf36b3", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caf765e5857f4bef49dfc87f2887d22feb9d9a0300ea9f5e8b649bd0cb8d6dd0", "generation_seconds_batch": 23.30500212602783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8577580451965332, "target_z": -0.23828130371212364, "ce": 5.831989288330078, "genuineness": 4.086852550506592, "identity": 0.6708255410194397, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f5ba6234103650e725017915809f167c10de698ede7b2e2e687600c2bcbf36b3.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "f632df815ec63078f32361ab18837d8f005368c616895fb9bc61dc400aa63381", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29bae39e2da7bf3b170373a56054cb2e82bcbc6cd14be0aca799ba4b24fae019", "generation_seconds_batch": 19.288109980989248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.780544638633728, "target_z": -0.14550786173611333, "ce": 5.8875885009765625, "genuineness": 3.6977379322052, "identity": 0.6940876245498657, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f632df815ec63078f32361ab18837d8f005368c616895fb9bc61dc400aa63381.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f649203481f8a45ce71d8a546149272c3efef56dcb298910e0febfab9006904d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "30c8eabc0c348e18033314819eab887a9e0c78b4557462c2ebfe99fa244d59b6", "generation_seconds_batch": 16.692984931985848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7025184631347656, "target_z": -0.05175786173611334, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.8235723972320557, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f649203481f8a45ce71d8a546149272c3efef56dcb298910e0febfab9006904d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "f6880ba99f05bbe53fa9b829da7b536319c79e99592fcb89216e21ff0ece2098", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0ad87873d5aafcf8ce40f35831dbe6a4278a971cab162ef66a4e3139be9b996e", "generation_seconds_batch": 18.182881890097633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7358421087265015, "target_z": -0.09179688395202061, "ce": 6.09409761428833, "genuineness": 4.494794845581055, "identity": 0.6376080513000488, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f6880ba99f05bbe53fa9b829da7b536319c79e99592fcb89216e21ff0ece2098.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "f6acaa9ee376471223895da26ce57c2b72f61085b0bc9f046402d5018f5d16de", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "382e67ec435d7da1953bbc1dbf06f1189afca3055b16a620119be3a5d660c9ac", "generation_seconds_batch": 15.882174622034654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4006900787353516, "target_z": -0.8906249283838351, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.6496310830116272, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f6acaa9ee376471223895da26ce57c2b72f61085b0bc9f046402d5018f5d16de.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f6c163273e649328fabf9284c61b94a02ad0fe3e4c7cfac22e60ed0a5a7e6716", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27fa9606e2c254933c4fdcca1fbf012db143fe837884b37a882828894b1d7242", "generation_seconds_batch": 21.940618842025287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5521554946899414, "target_z": 0.12890623209595878, "ce": 6.117925643920898, "genuineness": 4.494794845581055, "identity": 0.7504104375839233, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f6c163273e649328fabf9284c61b94a02ad0fe3e4c7cfac22e60ed0a5a7e6716.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "f6d5fffd605007b4f00b80f2b181b8279f2f15b0fb72b9c59d1aef71746ec334", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "48d59017b81677338a097c67230cf6907fd7ddc7fbe60cf5f16f487809283a82", "generation_seconds_batch": 19.189993995940313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9422863721847534, "target_z": -0.3398437679040412, "ce": 6.205295085906982, "genuineness": 3.7667744159698486, "identity": 0.35743775963783264, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f6d5fffd605007b4f00b80f2b181b8279f2f15b0fb72b9c59d1aef71746ec334.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "f71b86b3a4d7899a5a40b9da72e6b7aca6dc9b1ba5715629eec3114888bb4a91", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8cb4958a4fa43c48764f0f4751d60309fc9c3ebee0078e2b7f6b329a0466b3b5", "generation_seconds_batch": 18.946965656941757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6903268098831177, "target_z": -0.03710934814393818, "ce": 6.443574905395508, "genuineness": 4.469690799713135, "identity": 0.20434468984603882, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f71b86b3a4d7899a5a40b9da72e6b7aca6dc9b1ba5715629eec3114888bb4a91.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very relaxed and unstrained. 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": "8ac47c0ef794a4df66452f82423b6ff105fec4ed43d584d7a2f3d654bc966f69", "generation_seconds_batch": 18.10564396996051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6884981393814087, "target_z": -0.034912164206126216, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.7480443716049194, "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_TENS_reg_low/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "f81dfa4fc5fcdb913dfd6ef2156b5384de15acbdf51c7025473a2fc343587bb8", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "521df3fa9b6a4ddfdf7afa90a005daefe5c26ae1a3a755edeb84729e2e75944b", "generation_seconds_batch": 23.643021292053163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2186291217803955, "target_z": -0.6718750716161649, "ce": 6.062326908111572, "genuineness": 4.419482707977295, "identity": 0.7177892923355103, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f81dfa4fc5fcdb913dfd6ef2156b5384de15acbdf51c7025473a2fc343587bb8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "f828f16f770e2f4170308a818ab48ce268587a067d74e9b0e663e5c94afc67ec", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "0221e7a4be60fc3dfe02ce8636c33b4ad351f5e8b6f0e445d60f40c7b1634777", "generation_seconds_batch": 15.976094212965108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436741828918457, "target_z": 0.26757804443181454, "ce": 6.1894097328186035, "genuineness": 4.557555198669434, "identity": 0.7074254155158997, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f828f16f770e2f4170308a818ab48ce268587a067d74e9b0e663e5c94afc67ec.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "f858def5ef325adf495037deac72959cb6a9889de7ae81461731537e33a4cae5", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ec3325f57cabf7f1c504c4eba8d787a78b9a32c6b2fe4c5a4a2beb2d5b8ddd5", "generation_seconds_batch": 23.838397464016452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746814727783203, "target_z": -0.8593750716161649, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.4154609739780426, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f858def5ef325adf495037deac72959cb6a9889de7ae81461731537e33a4cae5.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_TENS_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very relaxed and unstrained. 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": "5c0b8d0fe97da3ad34b8c3b99c46ecac769c9c68e8cac565b464f8b88fb263f1", "generation_seconds_batch": 18.05504199594725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7260888814926147, "target_z": -0.08007818766414425, "ce": 6.395919322967529, "genuineness": 4.532451152801514, "identity": 0.4591529667377472, "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_TENS_reg_low/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.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": "f89dbbeb3795fd9239c600643a6424691a64e82f4de8760d7704e84a7fcc9d4e", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08ecb374ad08cb1c53edae04b9cd156e5f178e0bbd2c3cac1ec04a8236793034", "generation_seconds_batch": 24.595501671894453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5253340005874634, "target_z": 0.1611327901199485, "ce": 5.903473377227783, "genuineness": 4.645420074462891, "identity": 0.6287115812301636, "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_TENS_reg_low/f89dbbeb3795fd9239c600643a6424691a64e82f4de8760d7704e84a7fcc9d4e.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": "f8b962bc340a92060e7757e939a7fede64c3ca54647ee77e2dfb2a622058efb7", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "15484f2084757f713fcc18d2e627cbb823a1e3d3b42166f42b43941b31620719", "generation_seconds_batch": 18.74141849996522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7902978658676147, "target_z": -0.1572265580239897, "ce": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.7248789072036743, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f8b962bc340a92060e7757e939a7fede64c3ca54647ee77e2dfb2a622058efb7.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "f8d850f85af7f8a0ef7839685758b38761a3e9ac5d0fea8a24556a18148edd5b", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "25b319b06751f3a67253e0a7f41003e9e70e5faafc807b72837d982d7e13a527", "generation_seconds_batch": 19.770344026037492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.462460994720459, "target_z": -0.9648439111363709, "ce": 5.903473377227783, "genuineness": 4.193545341491699, "identity": 0.6432250738143921, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f8d850f85af7f8a0ef7839685758b38761a3e9ac5d0fea8a24556a18148edd5b.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-1", "key": "f92534557a6537f805363e04d03fafe9aa3bc32389b8ce11e63f8ed1780cf106", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "14c11d87f2840b2e4803816e1c00438293e4c616419146ca7eccb812b2a26a58", "generation_seconds_batch": 17.42860073398333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4383673667907715, "target_z": 0.26562492838383517, "ce": 6.054384231567383, "genuineness": 4.770940780639648, "identity": 0.7039585709571838, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f92534557a6537f805363e04d03fafe9aa3bc32389b8ce11e63f8ed1780cf106.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "f954899959ce7d6343706458ac4a1125ebb90b6394c815d9092552ddb0903e7d", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "181abb7d5ef8caef7e0942acacf4d7ac859c18068d7ca9b74e5761c8ee1cbe9c", "generation_seconds_batch": 24.132190837990493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6403412818908691, "target_z": 0.02294925679608758, "ce": 5.974957466125488, "genuineness": 4.20609712600708, "identity": 0.5789080858230591, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/f954899959ce7d6343706458ac4a1125ebb90b6394c815d9092552ddb0903e7d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "f960e2074f412a8d120290b6f26988d83c448bf61f909f7ee769bbd2d8386910", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "69c6c11d459e317a96e6d662429e28bd0f3438785ef2e79c5c725fd09353b12f", "generation_seconds_batch": 22.739485797006637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4318650960922241, "target_z": 0.27343753580808244, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.7263592481613159, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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_TENS_reg_low/f960e2074f412a8d120290b6f26988d83c448bf61f909f7ee769bbd2d8386910.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "fa0ecf8638f89e29c4ff7e74f7a9d2a63ca2fc63febfccdc6dc8d1343116142d", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "ff676ac542250060bdb058152b8358c69a9f81760441cb6e5c02244b7e05ea4e", "generation_seconds_batch": 15.243745477986522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6915459632873535, "target_z": -0.038574185179922726, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.5318042039871216, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fa0ecf8638f89e29c4ff7e74f7a9d2a63ca2fc63febfccdc6dc8d1343116142d.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": "fa53c253aaaffca39f7a241777d4d825cb1ebe3203134788a2071ef4e2b3bcf0", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The road turns left behind the bridge.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0169b16687cb01b18e1cb68b87f0bd4e5e0d9b4b5d19c742f0be6efaeb1d4a92", "generation_seconds_batch": 20.601640170090832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7301527261734009, "target_z": -0.08496097778409273, "ce": 6.41180419921875, "genuineness": 4.281409740447998, "identity": 0.30372607707977295, "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_TENS_reg_low/fa53c253aaaffca39f7a241777d4d825cb1ebe3203134788a2071ef4e2b3bcf0.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "fa69bd5b8ad7c04ef718961658e29e56694b40992250a93569e5896ead3a274c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3e99ad9ff82af39f6726c787c91120d9bb9677172b4d6e006af62f5ce89de4c", "generation_seconds_batch": 23.784617349971086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9325330257415771, "target_z": -0.32812492838383517, "ce": 6.006728172302246, "genuineness": 4.18099308013916, "identity": 0.6346685886383057, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fa69bd5b8ad7c04ef718961658e29e56694b40992250a93569e5896ead3a274c.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": "fb27e882b987c55b872adc5b4bb9abba7fbdddbe80eb0f5745ffd0448fc7de45", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0b0e0088af22875d8cdbcc305af0ee0221b8a50704d65ac42a170d87707543b8", "generation_seconds_batch": 24.237278033979237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5424022674560547, "target_z": 0.14062492838383514, "ce": 5.677107334136963, "genuineness": 4.570107460021973, "identity": 0.7550958395004272, "hypothesis": "J'écris le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.09090909090909091, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fb27e882b987c55b872adc5b4bb9abba7fbdddbe80eb0f5745ffd0448fc7de45.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-0", "key": "fb521d1f927995d83ce29ea12e19a0e18f85df700b83f55494e4158567a062f8", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b08e310078e2ffc951ba4732b5cc2b69e321c9804c22cd31b374cee6ead75a7", "generation_seconds_batch": 16.754659287980758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8008639812469482, "target_z": -0.16992195556818546, "ce": 6.165581703186035, "genuineness": 3.848362922668457, "identity": 0.3058433532714844, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fb521d1f927995d83ce29ea12e19a0e18f85df700b83f55494e4158567a062f8.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "fb947dde7682361918286528c1bce27f7d6b5c89e3b034fe0a9a64d2295a871f", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "857e91998c7f60e377c02eb1f2e338d6e52a46785a5e1e7a860b37b5b39254db", "generation_seconds_batch": 28.165058269980364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648469090461731, "target_z": 0.013183533323860902, "ce": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.5099259614944458, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fb947dde7682361918286528c1bce27f7d6b5c89e3b034fe0a9a64d2295a871f.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-en-1", "key": "fba34bfbd6f02958e0506d38dc16465ba6ff9d88c2d1070b1e4816a75734ae92", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "8ed57884347fa8a50c54ef8bcd718c8348ddc9cbd8df5b872507e8141b835085", "generation_seconds_batch": 16.72145465493668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.474129319190979, "target_z": 0.22265623209595878, "ce": 6.38797664642334, "genuineness": 4.469690799713135, "identity": 0.5198399424552917, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fba34bfbd6f02958e0506d38dc16465ba6ff9d88c2d1070b1e4816a75734ae92.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": "fbe18424b6d8c55158f82d0f37f9f83c5a36556f5c4eefc1db2e31b7f6f8e247", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [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": "e75d0e200eed3cd56743eba05c7536778aeb61c1e5a770b0ea875978a5a776ba", "generation_seconds_batch": 20.47187789203599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1861181259155273, "target_z": -0.6328124641919176, "ce": 6.213237762451172, "genuineness": 4.65797233581543, "identity": 0.3912706971168518, "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_TENS_reg_low/fbe18424b6d8c55158f82d0f37f9f83c5a36556f5c4eefc1db2e31b7f6f8e247.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "fc0181acac7fa98d860a4333e9a605113fb3d97327c58945fd353a2df99f5d49", "bucket": "vn_TENS_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": "bc7970828da41cfc17b89e69a3b4b27c994b28897e51768de6d14ae9bbf6c841", "generation_seconds_batch": 15.968356427038088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7317782640457153, "target_z": -0.08691409383207213, "ce": 6.419747352600098, "genuineness": 4.086852550506592, "identity": 0.5925331115722656, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fc0181acac7fa98d860a4333e9a605113fb3d97327c58945fd353a2df99f5d49.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "fc1684e3d1d989b23d44bd0005642a84227ab43821073ea3c25a8d145a27d94c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c73fd976a2aae188a60adb7ed58b33ebdea00d45d1c3d15d8869cd043bc09d1a", "generation_seconds_batch": 22.263895235024393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9731717109680176, "target_z": -0.3769531160479794, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.42841455340385437, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fc1684e3d1d989b23d44bd0005642a84227ab43821073ea3c25a8d145a27d94c.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-1", "key": "fd233ace09a5bd65b9dd88665bd5cb0e704cff44c35ef55aca947290797297c1", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b989165c6f7f44b956281769841add77a0f0d464d8457ecd1603f69a24b8875", "generation_seconds_batch": 23.43631063203793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4521844387054443, "target_z": 0.2490234419760103, "ce": 6.221180438995361, "genuineness": 4.946669578552246, "identity": 0.7520132064819336, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fd233ace09a5bd65b9dd88665bd5cb0e704cff44c35ef55aca947290797297c1.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "fd780ba6388d540c42cc70314d281aa9ad41a029796478d1cb22c07853f002c6", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "927ccd1aec5564e60d23631df24c1ca3bd8f8622b449f231a5e1accc2824b232", "generation_seconds_batch": 22.878705587936565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5643470287322998, "target_z": 0.11425786173611334, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.7945549488067627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fd780ba6388d540c42cc70314d281aa9ad41a029796478d1cb22c07853f002c6.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-1", "key": "fd9d04e91d3786e5e6cc95d68d5490cfaf000550afae0df82cf846c456f9ec71", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f429085bbca248e2e10f0eb0323dcee98b7e0da8f4ea2be6af8bcbb2172b0a7d", "generation_seconds_batch": 17.10025278897956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4838825464248657, "target_z": 0.21093753580808242, "ce": 6.467403411865234, "genuineness": 4.20609712600708, "identity": 0.5593421459197998, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fd9d04e91d3786e5e6cc95d68d5490cfaf000550afae0df82cf846c456f9ec71.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-es-0", "key": "fdb5c6a9b0b86a2c8c1786bc626261aa084cc69cd5c48e992f1792aae2687d48", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e08e2be2f7094e87b2ab8b1776cf2d2d3e76dc485bea2e18ff782100d5223b1", "generation_seconds_batch": 22.407986855017953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.791110634803772, "target_z": -0.1582031160479794, "ce": 6.006728172302246, "genuineness": 4.20609712600708, "identity": 0.7536991238594055, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fdb5c6a9b0b86a2c8c1786bc626261aa084cc69cd5c48e992f1792aae2687d48.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "fdb9375d663b12a0afb30296cea3ac4a8f841f1ec1723ee14371976a7ae441eb", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a484216cfedb663ab9e15f3c7a4b082f916cd0c439c6df2467b38fb29394c5de", "generation_seconds_batch": 17.27268993796315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.355175018310547, "target_z": -0.8359376790404122, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.5794615745544434, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fdb9375d663b12a0afb30296cea3ac4a8f841f1ec1723ee14371976a7ae441eb.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": "fde9c17c4b519fffeb9dfdfc6022958a7a378a5df0921b60e1866253a2f9593d", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4ec9799dbdf68e359a741fc5ebaca5d0b04d3e1ac5d59fd2633f9ae77a4cc099", "generation_seconds_batch": 23.91047299990896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7533167600631714, "target_z": -0.1127930247001288, "ce": 6.046441555023193, "genuineness": 4.570107460021973, "identity": 0.6389187574386597, "hypothesis": "De sleutels liggen op de keukentafel. Keken in de la en daarna naast het raam neem ze mee als je weg.", "wer": 0.13636363636363635, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fde9c17c4b519fffeb9dfdfc6022958a7a378a5df0921b60e1866253a2f9593d.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-0", "key": "fe0b3aeeafa4e38aac2f64917b683e50ea147a83b151142342b371f6c8924938", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a11a02d0924e3798a4610249f0751e9193d3753e46d59f19bbb28d972a515b0", "generation_seconds_batch": 23.643021292053163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8260598182678223, "target_z": -0.20019525431186605, "ce": 6.062326908111572, "genuineness": 4.375550270080566, "identity": 0.4725978970527649, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fe0b3aeeafa4e38aac2f64917b683e50ea147a83b151142342b371f6c8924938.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-de-0", "key": "fe0e19f7f8e565f9cbcd110ca5e8c4cb7b9a19256f1b3d26e4d758b7f02e6715", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2deaac7e9b19a41440f56dfa5df9627787ccd326f9dd97348f09a7d1ecd6b9fa", "generation_seconds_batch": 24.334042980102822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.550529956817627, "target_z": 0.13085934814393818, "ce": 5.720792293548584, "genuineness": 4.375550270080566, "identity": 0.5736621618270874, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/fe0e19f7f8e565f9cbcd110ca5e8c4cb7b9a19256f1b3d26e4d758b7f02e6715.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-fr-0", "key": "feaa9f4ff6f268b7edcbbfaa9e14c1796187f729b45c2097263f268835533087", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b9f34af967f5ad81a0f743ceb5abd39511438b3bf8876d40c5d211cc334e355", "generation_seconds_batch": 17.727184303919785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0547386407852173, "target_z": 0.7265624641919176, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.7444883584976196, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_low/feaa9f4ff6f268b7edcbbfaa9e14c1796187f729b45c2097263f268835533087.mp3" }, { "id": "vn_TENS_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_TENS_reg_low-nl-1", "key": "fec45f585d63346557d4cc0d1f841d17908193e7507c5276e2a0827c5862322c", "bucket": "vn_TENS_reg_low", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "54c84782dcf214f3f4fd35dc9a9111d187c9014b207b5a65617c7370ec72706b", "generation_seconds_batch": 19.687512636068277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6613718271255493, "target_z": -0.0023193611150579573, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.46191972494125366, "hypothesis": "Ik 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_TENS_reg_low/fec45f585d63346557d4cc0d1f841d17908193e7507c5276e2a0827c5862322c.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": "ffdd258bd1ce018a2a274bf492eee7294d2200ce5c145fe105c43048e7953a5a", "bucket": "vn_TENS_reg_low", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Audibly relaxed and unstrained delivery, with loose, comfortable voicing and easy articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9a304ca0d519c2372396f0cfb5a9a56be7318503a9ff39cb57a7141ba9bf4168", "generation_seconds_batch": 19.658320095972158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.091836452484131, "target_z": -0.519531160479794, "ce": 5.80021858215332, "genuineness": 4.582659721374512, "identity": 0.8044677376747131, "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_TENS_reg_low/ffdd258bd1ce018a2a274bf492eee7294d2200ce5c145fe105c43048e7953a5a.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "fe8ae9f9b440dd092acef122d903e6c374274db67a97010c3ae779cc3ab4190c", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.906524419784546, "target_z": -0.29687507161616483, "ce": 6.229123115539551, "genuineness": 4.494794845581055, "identity": 0.647192656993866, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/a4ec81561a48022ab3b0d002.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "b3a9f356e89a25310b3aa2d2c4644b88d2bed829a15d08a2cc8c28c261200c95", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 2.117845296859741, "target_z": -0.5507813037121236, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.5465888977050781, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/dd20d306f568907f57e38eb7.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "68e89621e8b7fdb682344fd098825a29e9703041c3aac27ab162effc5dc43d3b", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.6163644790649414, "target_z": 0.05175786173611334, "ce": 6.4515180587768555, "genuineness": 4.720732688903809, "identity": 0.5647629499435425, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/4bb5590f6e010cf78c87b1de.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "1aadbf49cfac9d95f79376c4dce1e7d99a0720839fb36dc0845d40d75209df6c", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.978048324584961, "target_z": -0.38281246419191756, "ce": 6.300607204437256, "genuineness": 4.268857479095459, "identity": 0.5856472253799438, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/feab5c2b743a50530ed97917.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "7c2f1b6a7b23822e4268c4657a1babf9f7d7a051d6b3920d12248b6192a033ae", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.994303822517395, "target_z": -0.4023437679040412, "ce": 6.292664527893066, "genuineness": 4.532451152801514, "identity": 0.7167495489120483, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/b0034673520794c0590352d6.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "6ae5f27e4e15726207e4d9e6d901c7ebdee8e8b48f5c73effc281d613fdee8c7", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.8512557744979858, "target_z": -0.23046869628787636, "ce": 6.149696350097656, "genuineness": 4.457139015197754, "identity": 0.7775472402572632, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/dcf2bae479bb2a30e34db8e2.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "baf8b7fcf0263491e22c2bd4b1b1fd958b0f1a79cc0990a97291b8f7e8be7d4d", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 1.8057405948638916, "target_z": -0.17578130371212364, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.43006664514541626, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/59ac63fecb2d7ff3f3b045a8.mp3" }, { "key": "original-vn_TENS_reg_low-de-0-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "3e6b2e2cf0ed998471fd5287173e02767f7480c73f7529cec57cfe7ecc47e6d2", "task_id": "vn_TENS_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very relaxed and unstrained. 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": 2.0788321495056152, "target_z": -0.5039062320959588, "ce": 6.0305562019348145, "genuineness": 4.607763767242432, "identity": 0.7528244256973267, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/409b8af8b01290c15435fb48.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "94fd7ce35f4512a9832bea012570b514081a0a6d95533287553eae28e2d6cb56", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.6086431741714478, "target_z": 0.061035162964015455, "ce": 6.213237762451172, "genuineness": 4.18726921081543, "identity": 0.5818609595298767, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/b173dea42719a990fcdd2b04.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "44a7c1e76e8647a0a9ed3150accdd996a5314a6b1ac3bfc234cfdc202cd0a998", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.670413851737976, "target_z": -0.013183533323860902, "ce": 6.268836498260498, "genuineness": 4.18099308013916, "identity": 0.565455436706543, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/b3fec9558fa0f2061deba6cb.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "f132f45901310aa37bb85c745c1d023aaa33b74743f2cf09b2ea55fb06be2d85", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.4879463911056519, "target_z": 0.20605474568813395, "ce": 6.435632705688477, "genuineness": 4.469690799713135, "identity": 0.6241186857223511, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/94e11270ec884b4020f6a528.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "2d14166aad07f654fe4a25f8de4222e570e7ce9ba036ef6aeeae2662208da65c", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.7512848377227783, "target_z": -0.11035162964015455, "ce": 6.3641486167907715, "genuineness": 4.13078498840332, "identity": 0.6597191691398621, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/0ff6381d5834f355cda3fb34.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "30048e15a18ced0a1b06dac93435f9491d9223dc6d659d5fece12d20be1844e1", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.6814879179000854, "target_z": -0.0264892438249678, "ce": 6.38797664642334, "genuineness": 4.394378662109375, "identity": 0.4004731774330139, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/840d396118cf4dd45c2a3820.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "2a8739ff235591d86a72937eb76fd3ab36aa8b82b222cb62702f20c33b4bca71", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.5980771780014038, "target_z": 0.0737304172758815, "ce": 6.395919322967529, "genuineness": 4.030367851257324, "identity": 0.44874316453933716, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/20e5789a87ce06926f490899.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "7bd8a3d0f77cbef21302228edf23774b660f9239974c4d8a8b73052c667b7017", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.6391221284866333, "target_z": 0.024414093832072124, "ce": 6.41180419921875, "genuineness": 4.331617832183838, "identity": 0.6416250467300415, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/de/4e52630bd7422540b76c50fd.mp3" }, { "key": "original-vn_TENS_reg_low-de-1-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "18cdbd597bcca42251c631f4f5968b5d4837022742d2f757f0bfef5efb269c6d", "task_id": "vn_TENS_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very relaxed and unstrained. 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": 1.7943617105484009, "target_z": -0.16210934814393818, "ce": 6.157639026641846, "genuineness": 4.356721878051758, "identity": 0.5909031629562378, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_low/de/c65c8b80c30de03e752e4b8c.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "0e51c3cb8efc609fed5292cf00032a3f50537b20c9d3d0a787ef14908f3af7f3", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.8098044395446777, "target_z": -0.18066409383207213, "ce": 6.181467056274414, "genuineness": 3.9550557136535645, "identity": 0.3510732650756836, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/dd822bce5a487143bac60465.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "8589f57b9c8b4b4f7e7d4a76046207b25b7bbecd76ac1a8ce52b2a9802e7759e", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.7707912921905518, "target_z": -0.13378902221590727, "ce": 6.324435234069824, "genuineness": 4.2751336097717285, "identity": 0.39723441004753113, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/f75953aebe7eacbe1611c705.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "b2425cf106f5961abd29c406f3ae007eb36705640dee66163037662220ac8466", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.7435634136199951, "target_z": -0.10107418517992273, "ce": 6.24500846862793, "genuineness": 4.074300289154053, "identity": 0.1343943476676941, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/916ec3538dda6f57e158e567.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "ff0adde94bcbc80fc903d4b5a8f8b1af989270726bd658d289fa22e16df8705d", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.7163355350494385, "target_z": -0.06835934814393818, "ce": 6.1020402908325195, "genuineness": 3.8044304847717285, "identity": 0.24078944325447083, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/df2bc086587f9d64779a1f0c.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "7991dde81af0e2a64a3d890d199092b1408b4801b80c74cc0c5c0711d9eabe84", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.8138682842254639, "target_z": -0.1855468839520206, "ce": 6.197352409362793, "genuineness": 4.055472373962402, "identity": 0.20524519681930542, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/a6d4e3dccd5ff6abe0124fd5.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "af6fb64fa6d2292b4d9295c178e1cc0b813b7c1e885831b65576c46d042887ce", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.689920425415039, "target_z": -0.03662106913194333, "ce": 6.3641486167907715, "genuineness": 4.162164688110352, "identity": 0.22744759917259216, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/e377dcb8c277d37dd9ec1a5a.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "7225b77f5f9b1949a4f64e9a3a6119cf246283925e7ea0f555f14eda475f2217", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.6078304052352905, "target_z": 0.06201172098800515, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.24853113293647766, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/caabc26bef26de3735d9977c.mp3" }, { "key": "original-vn_TENS_reg_low-en-0-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "54bea50c1a84408a1d905680f5d67edbe12febd983ec077869538a204b9513e6", "task_id": "vn_TENS_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very relaxed and unstrained. 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": 1.9878015518188477, "target_z": -0.3945311604797939, "ce": 6.213237762451172, "genuineness": 3.9299514293670654, "identity": 0.2908480763435364, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/188bf59a8972609d5063d025.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "066015bb082a230f7d1f1d3f2ee271c77e6246f8279b2d0ac2dcddd21db7dea2", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.5440278053283691, "target_z": 0.13867181233585574, "ce": 6.3641486167907715, "genuineness": 4.432034492492676, "identity": 0.5674779415130615, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/011deed4903b3ce7cd1e6d89.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "b1290fb7f49a9fdecea95e9aea1aaba701929bef20f560569e2ddfdbe83459f4", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.5440278053283691, "target_z": 0.13867181233585574, "ce": 6.41180419921875, "genuineness": 4.331617832183838, "identity": 0.49595826864242554, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/d00b8a7030eb4e141305c68c.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "cac09c12cd4e69ac0d88d35dfca791ee2a282c1ffb410cc7275863badde3e3ee", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.650501012802124, "target_z": 0.01074213826388666, "ce": 6.348263263702393, "genuineness": 4.65797233581543, "identity": 0.31890034675598145, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/1d884f1fdad52bfa11116090.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "4a967461515c3820bd0f5efac531f05a6ea15e5b25fa359328e1992b145eb49c", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.2790639400482178, "target_z": 0.4570311604797939, "ce": 6.395919322967529, "genuineness": 4.469690799713135, "identity": 0.33390384912490845, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/6ef0d4fc8165d76448137b81.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "36ebb8d382ce64f979bf07f2ea4f8f3952187f23d2f5fa882eea769056aecc2b", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.4091074466705322, "target_z": 0.30078130371212364, "ce": 6.3641486167907715, "genuineness": 4.457139015197754, "identity": 0.27010226249694824, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/fcd19f7bdc039cdf65385083.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "ca8a647c306d27968fac3921ec5a627bf4d8a4ed4733111b732d42020663e46d", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.481444239616394, "target_z": 0.2138672098800515, "ce": 6.284721851348877, "genuineness": 4.243753433227539, "identity": 0.3145138621330261, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/4f2265922725b8edfe2ab61f.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "751dd561cf7c1e6ba392cb11f3301bad8363543764ab15d5f5c93fe391b954fc", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.5757259130477905, "target_z": 0.10058590616792787, "ce": 6.38003396987915, "genuineness": 4.20609712600708, "identity": 0.3569871187210083, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/d323d7c3a4107cdeea895f14.mp3" }, { "key": "original-vn_TENS_reg_low-en-1-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "97d90f9856dfbf2bc250df6af0e26c2ea8506801b9067147cda5570f9957215f", "task_id": "vn_TENS_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very relaxed and unstrained. 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": 1.6076271533966064, "target_z": 0.06225593211016743, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.4823150336742401, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/en/de96249f32f288f3a8290984.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "3c5456bcc133bf9b3d93221b4e03c7d602a558120c44282b7d9ce429888cb97c", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.709426999092102, "target_z": -0.06005860494002576, "ce": 6.117925643920898, "genuineness": 4.068024158477783, "identity": 0.4617568850517273, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/16b8cbcbb2e2ee2359f0e4b7.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "716315224149a1df5142c2e22109cb49cc74c5c694a2a2c93f500e1e6d969f8d", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.6947970390319824, "target_z": -0.04248041727588151, "ce": 6.078212261199951, "genuineness": 4.13078498840332, "identity": 0.4750397801399231, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/1dc602b7424778cbc5910f78.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "e9c76ad94e2e0cd22e3f3819aefe4c5de42eff90f6648452ae7c5fbacc195131", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.5911685228347778, "target_z": 0.08203130371212364, "ce": 6.213237762451172, "genuineness": 4.287685871124268, "identity": 0.49841827154159546, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/cfa199c022e57b9d43eba371.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "89dedba3c505d2079097833dfdc7da53d67714378d6107428666f626173d90f5", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.6123006343841553, "target_z": 0.05664065185606182, "ce": 5.919358730316162, "genuineness": 3.7416701316833496, "identity": 0.4178815484046936, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/aba3a3cbfd90c5cfa8425605.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "1c5dcc5d877e5b20530cbf4396bc561effc12dbc2a7bf119ce697d91a3a99181", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.7358421087265015, "target_z": -0.09179688395202061, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.5851119160652161, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/9fb970389b20ae4e28658313.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "d181a7639afabc9c0d8017fab2fb7234bd593d782e8c051b182c3c99564478e1", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.6496882438659668, "target_z": 0.011718696287876357, "ce": 5.8875885009765625, "genuineness": 3.9299514293670654, "identity": 0.5112961530685425, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/8965022b306bc19a0eb24d5e.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "225b7657b10955d7045ac316822fec4cacb2bfeeaa1857d591042751c78e119f", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.6667563915252686, "target_z": -0.008789022215907268, "ce": 5.919358730316162, "genuineness": 4.21237325668335, "identity": 0.44588178396224976, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/fbf40a6a6a3e353ef5eb6afd.mp3" }, { "key": "original-vn_TENS_reg_low-es-0-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "0e90c3fdc64f30745628773340ce9142faf18aad7f8fcf4a10911c3bdba05396", "task_id": "vn_TENS_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very relaxed and unstrained. 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": 1.591981291770935, "target_z": 0.08105474568813395, "ce": 5.824047088623047, "genuineness": 4.042920112609863, "identity": 0.46987977623939514, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/c0592979a48176afa046e85a.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "2c13b11ccaf247ffbef33da4eac42c982c64979a9cdcc0ff9d7264c46dbced2f", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.7110525369644165, "target_z": -0.06201172098800515, "ce": 6.038498878479004, "genuineness": 4.65797233581543, "identity": 0.6828416585922241, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/45183e897fe0f719c82c871e.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "23a6aa7ae16e5e62b9cc6406cb6bdec4b2ada274be3db8d49e65539211b6133e", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.2237952947616577, "target_z": 0.5234375358080824, "ce": 6.117925643920898, "genuineness": 4.375550270080566, "identity": 0.5985385179519653, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/e549a3aa324dc78a069f3b8d.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "c0cacc1fe348c58ad4df70e59cf499f62c838c1936ec99823490471365e5be53", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.810617208480835, "target_z": -0.18164065185606182, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.6796524524688721, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/8930b406422eb0c76773db78.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "1e831d58cec3fc64f2511a8aaa963274044cdf6cbf269b3d61af77c23a0c2f13", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.4749420881271362, "target_z": 0.2216796740719691, "ce": 6.149696350097656, "genuineness": 4.582659721374512, "identity": 0.6062028408050537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/bedbd680371f1133631490f6.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "98b2361784f8070c2eb3fd06268c791283ba9c19045d3a616c45b209db45ad95", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.737467646598816, "target_z": -0.09375, "ce": 6.157639026641846, "genuineness": 4.7081804275512695, "identity": 0.7500888705253601, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/df854848fc49e52399b414fb.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "bc1b52d970b6364d8a48c1cb925b71504002e60f30c45059ef3925809ef07907", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.3961031436920166, "target_z": 0.3164062320959588, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.5893990993499756, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/ce4b541f3747221e69d9e30e.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "666dd00271821fa34b6608460706b38af89e869e5a717296269c477c0f468e9d", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.6517201662063599, "target_z": 0.009277301227902116, "ce": 6.1020402908325195, "genuineness": 4.381826400756836, "identity": 0.48140090703964233, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/256898f80c976150258dfb47.mp3" }, { "key": "original-vn_TENS_reg_low-es-1-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "ce6bfcba075debc5679dc0f12012ff2ed973a0621b781b6a1c521b02a1db631c", "task_id": "vn_TENS_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very relaxed and unstrained. 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": 1.8553197383880615, "target_z": -0.23535162964015455, "ce": 5.879645347595215, "genuineness": 4.7458367347717285, "identity": 0.6483808755874634, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/es/41df82b67162f1f2a573ae74.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "3ae2972488ee05907efa36c7ba63f0bde7d4bab2117ea8609b4ceff0fda76776", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.60214102268219, "target_z": 0.06884762715593302, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.7239694595336914, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/8952beb257aa38ec08639ad3.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "343b686245d4ecd3b98be1b3405744e81b59b1b1a6bbafc0ac7f691720681870", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.6545648574829102, "target_z": 0.0058593481439381786, "ce": 6.181467056274414, "genuineness": 4.356721878051758, "identity": 0.6192827224731445, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/da51c430568d4a66cdec2dca.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "9f94debae5edcb29bdd3aa970c28b1e9d3f0a811669542d94e3d0ae51f06c853", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.6744778156280518, "target_z": -0.0180664666761391, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.5979019403457642, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/6d012d435ddfaff40ce4408e.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "2a63092ad5f22875210c0ff574fc3b1c9143ac64e72509986f51462a0b8fb633", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.383098840713501, "target_z": 0.3320311604797939, "ce": 6.054384231567383, "genuineness": 4.13078498840332, "identity": 0.6377514004707336, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/c3a1c2da1ca7caaacbc438c4.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "605548db22a39c7bd83adcf7d5d1ba17906517f57ea19704bcca2937b7ae8581", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.6090495586395264, "target_z": 0.060546883952020604, "ce": 6.1894097328186035, "genuineness": 4.3127899169921875, "identity": 0.6096000671386719, "hypothesis": "La porte étant 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/audio/vn_TENS_reg_low/fr/3f68eae22e9c988bfb274e90.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "daee576cf13e8d23fe7cf3d666013a70e68f172974d8f6a1142e8d3e9878c69a", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.5090785026550293, "target_z": 0.18066409383207213, "ce": 6.332377910614014, "genuineness": 4.519899368286133, "identity": 0.7972902059555054, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/56578904a05a7790a7084e7d.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "383fc9c511d727d6e78fca77fd6a1726421f606c0bb1fe65fc3e7d554f0510b6", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.527772307395935, "target_z": 0.1582031160479794, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.8412106037139893, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/c844a55528b1d19a4732a5fd.mp3" }, { "key": "original-vn_TENS_reg_low-fr-0-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "5a18eb31670329709410c206ecb54b70054c76ddfe1aefe6a8dd78261eb32799", "task_id": "vn_TENS_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very relaxed and unstrained. 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": 1.6293689012527466, "target_z": 0.03613279011994848, "ce": 6.09409761428833, "genuineness": 4.162164688110352, "identity": 0.6843593120574951, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/0f7004fd165b038d847fc654.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "6a45d831eb9131ae73fe0063440d73727e408e7a32218d72ead703e46a09075f", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.4895720481872559, "target_z": 0.20410148640782483, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.7506797313690186, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/9f211499ab1317b7f3734d9b.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "d6bdc4bffb15d35c0d5d6ea5781912b4476b126cc0f56b73c6e0dd5cae3e07ff", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.6387157440185547, "target_z": 0.02490237284406697, "ce": 6.062326908111572, "genuineness": 4.595211505889893, "identity": 0.837086021900177, "hypothesis": "J'ai trouvé la petite boîte. Elle était très litée 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/audio/vn_TENS_reg_low/fr/0fbe4a1b72e4d769c2df7bed.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "720257e53669db601a3947bf8b9e4655aeb5eecadecce676cb29bdb3c5590ade", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.4985125064849854, "target_z": 0.19335934814393818, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.6654971837997437, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/6351f882882ab8d4fda32aff.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "c6974731a182bd6c5ffb07ee722c30eecd814f05b815b7d150b1091b2eca9e8b", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.582634449005127, "target_z": 0.09228516296401545, "ce": 6.086154937744141, "genuineness": 4.344170093536377, "identity": 0.6445379257202148, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/9b890d2e9ce1ac865e4d1ae3.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "fa0977976c84512c8b719fa906294b41aae14ae7580402045bff5d3098c156fc", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.5196446180343628, "target_z": 0.16796869628787636, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.7099660634994507, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/d40e00560081c3692afdff3d.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "4b20e24d8b4715b4d5fd7c52a6d5a64b67f578807f65eed4057e4855b0da1381", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.4570610523223877, "target_z": 0.24316409383207213, "ce": 6.054384231567383, "genuineness": 4.419482707977295, "identity": 0.8013167977333069, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/cd67410ceb68ce527345a091.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "69e2406706ba0d8fc67f8266faccff572a4a0c89a0f2c4e57bc9841633a82355", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.5025763511657715, "target_z": 0.1884765580239897, "ce": 6.213237762451172, "genuineness": 4.394378662109375, "identity": 0.7220004796981812, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/cfc7944aad64bb4ce4e11b91.mp3" }, { "key": "original-vn_TENS_reg_low-fr-1-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "bb558c7fb5f65af2108c0ca509c1b9d3292c73c8732ed65981db12505a16db77", "task_id": "vn_TENS_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very relaxed and unstrained. 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": 1.6088464260101318, "target_z": 0.06079095184185317, "ce": 6.149696350097656, "genuineness": 4.030367851257324, "identity": 0.7353565692901611, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/fr/8d988e8d9f0cf469051b1825.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "3cbc8221d88d4921803678bee6859185ae108c532436f1017f1ea6868d2d3f77", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. 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": 2.1145942211151123, "target_z": -0.5468750716161649, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.5249897241592407, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/e52a66753535d953c7076aa1.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "e5d31e19aba5b010b3bac29e266a5d2059b2391a5f74616cf65bda683a0e38b6", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9162776470184326, "target_z": -0.3085937679040412, "ce": 5.982900142669678, "genuineness": 4.425758361816406, "identity": 0.467180073261261, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/897a5a6428ae9aa32b24b1d4.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "9f15638f920ce4a04fe6b6e67860e9103260d2b53009a659ea9e315550bc440d", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. 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": 2.0398190021514893, "target_z": -0.4570311604797939, "ce": 6.109982967376709, "genuineness": 4.545003414154053, "identity": 0.48011282086372375, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/9a7053819cd3da4d078e90b9.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "d3760e649eebaebdff1acf01c73844a88eb84b4fb8c84e9ac42c1dd20a0ef1c6", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.898396611213684, "target_z": -0.2871093481439382, "ce": 6.09409761428833, "genuineness": 4.024092197418213, "identity": 0.11608821153640747, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/fcc8d54773b3728c7e40d445.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "fdf6cb6d9c0412b441c58d7356c14cc00f0d53f2b363fce51bf7e082aaa32bcf", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. 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": 2.127598524093628, "target_z": -0.5625, "ce": 6.0146708488464355, "genuineness": 4.400654315948486, "identity": 0.7198662161827087, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/0f66d55b4a3ce194af1b724b.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "8bdae35853fef8f494cf2cb769d7c640be50ed5a0056dcdeb20da4472a1058cb", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.88214111328125, "target_z": -0.26757804443181454, "ce": 6.173524379730225, "genuineness": 4.419482707977295, "identity": 0.6158475279808044, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/f1a113acc394211199b77630.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "03fabedeb253b66530364ae356eca15af299f14f88efa2720d9b73b14cc4bb96", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9552907943725586, "target_z": -0.3554688395202061, "ce": 6.078212261199951, "genuineness": 4.356721878051758, "identity": 0.7611728310585022, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/bb78eacce1aff00d7f68caca.mp3" }, { "key": "original-vn_TENS_reg_low-nl-0-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "bbc793804d66f9d5f4cb86b5941ab968957dbd6a2a9bf82022b15ae4304befc6", "task_id": "vn_TENS_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very relaxed and unstrained. 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": 2.1893692016601562, "target_z": -0.6367186962878764, "ce": 5.903473377227783, "genuineness": 4.168440818786621, "identity": 0.6771351099014282, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/1567c92b2b719a15d03ccf56.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1701-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "b8f953ad340b8479cf4eb6a81c2d27a980719fab679cfdae7e52d177ee986689", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.1341006755828857, "target_z": -0.5703124641919176, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.662887454032898, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/fb73c18f6cf605dc58335e2d.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1701-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "62a39dbc7b93c80930328dd7c1882a8f05473d2c6210c45d670a7a6b8826f855", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.1080920696258545, "target_z": -0.5390626074242473, "ce": 6.221180438995361, "genuineness": 4.425758361816406, "identity": 0.6960208415985107, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/a3f2dc4a862b36f4c0251378.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1701-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "05ca8cfdf91e58b58189cf1164c756d76d2dae44f9bd37f27aa9c5d8097a7406", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.067453384399414, "target_z": -0.490234419760103, "ce": 6.038498878479004, "genuineness": 4.482243061065674, "identity": 0.6476472616195679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/c4d80b87d3da62a822792990.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1701-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "f475d7011cf88cc58a44b57ad01f783fd792392b86cb9cf46089de467dfff378", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.1373517513275146, "target_z": -0.5742186962878764, "ce": 6.133810997009277, "genuineness": 3.9801597595214844, "identity": 0.6528811454772949, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/3532f149453b9c3addd3c173.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1702-scale0.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "3c5a5d10dfea9dbc66710a68ce088fa90cd5484108dcec3f712e75c5768342f3", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 1.8228087425231934, "target_z": -0.19628902221590727, "ce": 6.221180438995361, "genuineness": 4.444586753845215, "identity": 0.6482366919517517, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/bc5117993377faa8af7f022f.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1702-scale0.5", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "c6bb406c7ef54e94533e0f63d98ee168cf6f727f7fab0c9b09a3018eb62fa073", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.34867262840271, "target_z": -0.8281249283838351, "ce": 6.046441555023193, "genuineness": 4.231201171875, "identity": 0.6710256338119507, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/23474259f48d25f61d22ce14.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1702-scale0.75", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "05b53cc556eae1d7f6fd985aa12539b77fb30fa88d4890be8ff83b7f869e4cfb", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.1048407554626465, "target_z": -0.5351560888636291, "ce": 6.213237762451172, "genuineness": 4.469690799713135, "identity": 0.7625348567962646, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen het er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_low/nl/8b599119ccf84d6714fca688.mp3" }, { "key": "original-vn_TENS_reg_low-nl-1-seed1702-scale1.0", "bucket": "vn_TENS_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. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "90c9705d7d6c838eaedf4baba6edaf85da4ba0e036fa957f05b479580cbf0113", "mp3_sha256": "649f2b865950b08cf8986caab0adf77557afd1b07ea480c09431f2646260b7c0", "task_id": "vn_TENS_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very relaxed and unstrained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very relaxed and unstrained. 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": 2.2446377277374268, "target_z": -0.7031249283838351, "ce": 6.125868320465088, "genuineness": 4.331617832183838, "identity": 0.7133299708366394, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_TENS_reg_low/nl/4b1d250c0e27a1eb80d4c615.mp3" } ], "source_standard_audio_unchanged": true }