{ "complete": true, "report": { "bucket": "vn_TENS_reg_high", "label": "very tense and strained", "definition": { "id": "vn_TENS_reg_high", "category": "bipolar", "head": "vn_TENS_reg", "pole": "high", "label": "very tense and strained", "floor": 4.0, "canonical_dimension_name": "Tension" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P02", "general": "Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "inline": [ "Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0 }, "n": 20, "target_z": 1.0947631610458481, "target_raw": 2.5705900847911836, "ce": 6.170744466781616, "genuineness": 4.451490318775177, "identity": 0.5651575371623039, "wer": 0.016666666666666666, "valid_fraction": 0.95, "identity_delta": -0.04750653803348541, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.5084472842004428, "target_raw": 2.914890778064728, "ce": 5.8919566631317135, "genuineness": 4.255050265789032, "identity": 0.5773281797766685, "wer": 0.013157894736842105, "valid_fraction": 0.95, "identity_delta": -0.035335895419120786, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.0947631610458481, "target_raw": 2.5705900847911836, "ce": 6.170744466781616, "genuineness": 4.451490318775177, "identity": 0.5651575371623039, "wer": 0.016666666666666666, "valid_fraction": 0.95, "identity_delta": -0.04750653803348541, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.0084472627155934, "target_raw": 2.4987511575222014, "ce": 6.174318647384643, "genuineness": 4.468121886253357, "identity": 0.6419048741459846, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.029240798950195313, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.6494018546294981, "target_raw": 2.1999251306056977, "ce": 6.127059745788574, "genuineness": 4.452745580673218, "identity": 0.5711459964513779, "wer": 0.01990862573099415, "valid_fraction": 0.95, "identity_delta": -0.04151807874441147, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 0.5686279445702342, "target_raw": 2.1326986849308014, "ce": 6.190998220443726, "genuineness": 4.477535939216613, "identity": 0.5762418553233146, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.03642221987247467, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 0.493249510431897, "target_raw": 2.0699627816677095, "ce": 6.223563218116761, "genuineness": 4.43391740322113, "identity": 0.5728493109345436, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03981476426124573, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 0.49246826401270527, "target_raw": 2.0693125665187835, "ce": 6.189012575149536, "genuineness": 4.417599821090699, "identity": 0.5747548326849937, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.037909242510795596, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.1912109491376266, "target_raw": 3.483140778541565, "ce": 5.894339537620544, "genuineness": 4.136746907234192, "identity": 0.5193391297012567, "wer": 0.01488970588235294, "valid_fraction": 0.95, "identity_delta": -0.09332494549453259, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.8324218696287877, "target_raw": 3.184528088569641, "ce": 6.005238962173462, "genuineness": 4.2770164728164675, "identity": 0.544790980219841, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.06787309497594833, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.5084472842004428, "target_raw": 2.914890778064728, "ce": 5.8919566631317135, "genuineness": 4.255050265789032, "identity": 0.5773281797766685, "wer": 0.013157894736842105, "valid_fraction": 0.95, "identity_delta": -0.035335895419120786, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.2666992138263886, "target_raw": 2.713688886165619, "ce": 6.08218367099762, "genuineness": 4.363625764846802, "identity": 0.590926493704319, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": -0.021737581491470336, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.8751709048158147, "target_raw": 2.387828016281128, "ce": 6.080198001861572, "genuineness": 4.400968217849732, "identity": 0.5739519387483597, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03871213644742966, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.8536132718503784, "target_raw": 2.3698860466480256, "ce": 6.1131600618362425, "genuineness": 4.434231233596802, "identity": 0.6036983653903008, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.008965709805488586, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.7498535306196346, "target_raw": 2.283528971672058, "ce": 6.153667688369751, "genuineness": 4.446469569206238, "identity": 0.5774363473057746, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.035227727890014646, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 0.5821533304954234, "target_raw": 2.1439555823802947, "ce": 6.160816121101379, "genuineness": 4.448666119575501, "identity": 0.6049940660595894, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.007670009136199951, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 0.4539795025942239, "target_raw": 2.037279170751572, "ce": 6.144930768013, "genuineness": 4.400026845932007, "identity": 0.573915533721447, "wer": 0.008187134502923975, "valid_fraction": 1, "identity_delta": -0.03874854147434235, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 0.2898651166827851, "target_raw": 1.9006901800632476, "ce": 6.168163084983826, "genuineness": 4.399399197101593, "identity": 0.5823426231741905, "wer": 0.01111111111111111, "valid_fraction": 0.95, "identity_delta": -0.030321452021598815, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 2.7105469125984865, "target_raw": 3.9153733015060426, "ce": 5.640372610092163, "genuineness": 3.961802327632904, "identity": 0.5129810020327568, "wer": 0.009027777777777777, "valid_fraction": 1, "identity_delta": -0.09968307316303253, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.5582031160479795, "target_raw": 3.7885807275772097, "ce": 5.399473810195923, "genuineness": 3.9210080027580263, "identity": 0.514822892099619, "wer": 0.014791666666666667, "valid_fraction": 0.95, "identity_delta": -0.09784118309617043, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 2.45625, "target_raw": 3.7037272691726684, "ce": 5.660328578948975, "genuineness": 4.044489097595215, "identity": 0.521730725467205, "wer": 0.010818713450292398, "valid_fraction": 1, "identity_delta": -0.09093334972858429, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 2.3050780874015135, "target_raw": 3.577910029888153, "ce": 5.589936590194702, "genuineness": 4.051706552505493, "identity": 0.5582965672016144, "wer": 0.02900735294117647, "valid_fraction": 0.95, "identity_delta": -0.05436750799417496, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 2.1958984133295445, "target_raw": 3.4870420575141905, "ce": 5.696269154548645, "genuineness": 4.10724949836731, "identity": 0.5476715132594109, "wer": 0.008333333333333333, "valid_fraction": 1, "identity_delta": -0.06499256193637848, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 2.1199219125984867, "target_raw": 3.4238083958625793, "ce": 5.922734498977661, "genuineness": 4.169696164131165, "identity": 0.5592274658381939, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": -0.05343660935759544, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.7550781375328288, "target_raw": 3.1201565086841585, "ce": 5.731887030601501, "genuineness": 4.187896621227265, "identity": 0.5476129844784736, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.06505109071731567, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.4354980417275882, "target_raw": 2.854176640510559, "ce": 6.032541966438293, "genuineness": 4.337894010543823, "identity": 0.550757358968258, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.06190671622753143, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.1773437249343424, "target_raw": 2.639320170879364, "ce": 6.026882791519165, "genuineness": 4.37335364818573, "identity": 0.5600685834884643, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.052595491707324984, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.0549804888920464, "target_raw": 2.5374797940254212, "ce": 6.00077121257782, "genuineness": 4.374295043945312, "identity": 0.5466023579239845, "wer": 0.01067251461988304, "valid_fraction": 1, "identity_delta": -0.0660617172718048, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 0.5963867353112381, "target_raw": 2.1558017492294312, "ce": 6.142945075035096, "genuineness": 4.410382318496704, "identity": 0.5493161782622338, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.06334789693355561, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 0.5761718624671711, "target_raw": 2.1389773309230806, "ce": 6.188615441322327, "genuineness": 4.377746903896332, "identity": 0.5601099401712417, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.052554135024547574, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 0.480541973947758, "target_raw": 2.0593865633010866, "ce": 6.160418963432312, "genuineness": 4.401282024383545, "identity": 0.5576128505170346, "wer": 0.017076023391812866, "valid_fraction": 0.95, "identity_delta": -0.055051224678754805, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 0.457763679260417, "target_raw": 2.0404286623001098, "ce": 6.208472108840942, "genuineness": 4.401595783233643, "identity": 0.5618100628256798, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.05085401237010956, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 0.45217285592661005, "target_raw": 2.0357755362987517, "ce": 6.179084253311157, "genuineness": 4.421992993354797, "identity": 0.5624441459774971, "wer": 0.02, "valid_fraction": 0.95, "identity_delta": -0.050219929218292235, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 0.3868896583966229, "target_raw": 1.9814416885375976, "ce": 6.208074998855591, "genuineness": 4.424503374099731, "identity": 0.5568897157907486, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.05577435940504074, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 0.9752685606182137, "target_raw": 2.47113721370697, "ce": 6.163596057891846, "genuineness": 4.270112776756287, "identity": 0.6126640751957894, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 0.9691406088863629, "target_raw": 2.4660370469093325, "ce": 6.196955251693725, "genuineness": 4.29239274263382, "identity": 0.5767709493637085, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03589312583208084, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.0448241923415391, "target_raw": 2.5290269196033477, "ce": 6.159624695777893, "genuineness": 4.1800516724586485, "identity": 0.5583493933081627, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.05431468188762665, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.0526977522836962, "target_raw": 2.5355799198150635, "ce": 6.1439379215240475, "genuineness": 4.192917454242706, "identity": 0.5877522468566895, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.024911828339099884, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 0.9297656346681823, "target_raw": 2.4332660746574404, "ce": 6.235477237701416, "genuineness": 4.305760746002197, "identity": 0.5616777104139328, "wer": 0.0037267080745341614, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.5208593882489905, "target_raw": 2.92522111415863, "ce": 6.147154722213745, "genuineness": 4.31981906414032, "identity": 0.5567243963479995, "wer": 0.021119968894568436, "valid_fraction": 0.96, "identity_delta": -0.004953314065933228, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.6910937736333345, "target_raw": 3.0669036531448364, "ce": 6.020052108764649, "genuineness": 4.215134692192078, "identity": 0.579901976287365, "wer": 0.011609260304912479, "valid_fraction": 1, "identity_delta": 0.01822426587343216, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.5208593882489905, "target_raw": 2.92522111415863, "ce": 6.147154722213745, "genuineness": 4.31981906414032, "identity": 0.5567243963479995, "wer": 0.021119968894568436, "valid_fraction": 0.96, "identity_delta": -0.004953314065933228, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.1702343853843439, "ci95_z": [ 0.07249998854141362, 0.2731250429696989 ], "language_deltas": { "en": 0.07734373925757528, "de": 0.2828125214848495, "fr": 0.051953116047979385, "nl": 0.10390623209595881, "es": 0.3351563180353566 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.12710261344909668, "delta_genuineness": -0.10468437194824219, "delta_wer": -0.009510708589655957, "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": 0.1702343853843439, "ci95_z": [ 0.07249998854141362, 0.2731250429696989 ], "language_deltas": { "en": 0.07734373925757528, "de": 0.2828125214848495, "fr": 0.051953116047979385, "nl": 0.10390623209595881, "es": 0.3351563180353566 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.12710261344909668, "delta_genuineness": -0.10468437194824219, "delta_wer": -0.009510708589655957, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.5910937535808083, "ci95_z": [ 0.45425779011994855, 0.7213281418297987 ], "language_deltas": { "en": 0.6382812464191918, "de": 0.37519526863509906, "fr": 0.7962891081553051, "nl": 0.6878906232095958, "es": 0.4578125214848495 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.0883225154876709, "delta_genuineness": 0.014058318138122559, "delta_wer": 0.017393260820034274, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": true, "scale": 0.75, "cfg": 3.0, "reason": "Held-out target improves against matched CFG and best manual-only configuration; QA passes" }, "reference_only": true, "updated": 1793036104.4957328, "example": { "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.", "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." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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.", "language": "fr", "seed": 1702, "scale": 0.75, "cfg": 3.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "000b6b495dbe9324ed9c2176f1f833faf2d915e32632c37b4f4d2618ad3d8e95", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ab6e80eddb22109cf49d296bb5e04001167bb7478bd4c33c1a706e4876d1dc94", "generation_seconds_batch": 24.025337156956084, "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": 6.443574905395508, "genuineness": 4.394378662109375, "identity": 0.5046023726463318, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/000b6b495dbe9324ed9c2176f1f833faf2d915e32632c37b4f4d2618ad3d8e95.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": "00378963345ee3404bac83597eb9c7547d78dab15f1d848ecacf34bd000c2294", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cae0470eb0415063b2bd6af4cd0199e8d48c8e6962572b5f2255a3c9e22afa4b", "generation_seconds_batch": 16.665557779022492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.395919322967529, "genuineness": 4.036643981933594, "identity": 0.31694477796554565, "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_high/00378963345ee3404bac83597eb9c7547d78dab15f1d848ecacf34bd000c2294.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "005ba6b2543c7539849e2f9fa0ac61b539ce8857cea4d8795538651f41ae2322", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cbcdc9a5df2caf446a1190ff876b9ee2c3d024330f05ce9440357b8b95b959ce", "generation_seconds_batch": 18.092408030992374, "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": 6.038498878479004, "genuineness": 4.6956281661987305, "identity": 0.6341123580932617, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/005ba6b2543c7539849e2f9fa0ac61b539ce8857cea4d8795538651f41ae2322.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "006f4f11cf32a75b713b0c4a5313d5ad4eceaa3498c42b560131741eacbeb860", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "821dafcfc688b885cf8f6754ef8bb8e6184d66bf716fe2147368c70c512fbe3b", "generation_seconds_batch": 22.356071872985922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0834193229675293, "target_z": 1.710937679040412, "ce": 4.928510665893555, "genuineness": 3.9676074981689453, "identity": 0.5763713121414185, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/006f4f11cf32a75b713b0c4a5313d5ad4eceaa3498c42b560131741eacbeb860.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "00825242f00f37c5f94698ec2f5167584b7445e91669e32a4744048a3d943f42", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3e3a660f0ddbe6b5263ce9d40a13b6e981e2326b69cfe462bce45e8462120cc3", "generation_seconds_batch": 17.628292661043815, "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": 6.419747352600098, "genuineness": 4.469690799713135, "identity": 0.5446778535842896, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/00825242f00f37c5f94698ec2f5167584b7445e91669e32a4744048a3d943f42.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "014a3e793f134f6416d6ed56b42203dac884b71bc57da3ca8a7c8797cd2552f7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ebdd786728c480a3a1a833992a1c5d5e5be501f8564e2088776f781cb6b5f223", "generation_seconds_batch": 17.206736281048506, "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": 6.09409761428833, "genuineness": 4.7458367347717285, "identity": 0.7344871759414673, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/014a3e793f134f6416d6ed56b42203dac884b71bc57da3ca8a7c8797cd2552f7.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": "014bb455a86abf07f156778e12f86fbf7a3f9a5a5fdf830379f490d397dd8ba4", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5bc5c1f23c775780cfa1cafe5c2cadf17eebadcfef1087bdc819fb7c479745b", "generation_seconds_batch": 16.94114570110105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.455958604812622, "target_z": 0.957031160479794, "ce": 6.141753673553467, "genuineness": 4.770940780639648, "identity": 0.5302003622055054, "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_high/014bb455a86abf07f156778e12f86fbf7a3f9a5a5fdf830379f490d397dd8ba4.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "017a5356f77679a3ceab4bd771e5d5546603a3e2c936fb55f80dc766a7cfe81c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ed71ff8baf7206d6f5592be92c8b156111b16cb4cc6dbee17729b17bebcdca29", "generation_seconds_batch": 18.612191164982505, "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": 5.8875885009765625, "genuineness": 4.632867813110352, "identity": 0.5886857509613037, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/017a5356f77679a3ceab4bd771e5d5546603a3e2c936fb55f80dc766a7cfe81c.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "01961733792797c20dc3f916e19e1bb0f2e0b5880b2c236da6279e6ff472ed61", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "20943af5dc9b4ac2bf1ce59c61ebc96cd91798a8a91f1e6527c4ffb69af6f542", "generation_seconds_batch": 17.701858429936692, "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": 6.419747352600098, "genuineness": 4.457139015197754, "identity": 0.4037761390209198, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/01961733792797c20dc3f916e19e1bb0f2e0b5880b2c236da6279e6ff472ed61.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "bdfb32e5c9da62cc0cf82edd8290d4147feb92f24409c299b6e4f6aed66aaa1b", "generation_seconds_batch": 15.875765564967878, "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": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.7425732612609863, "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_high/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "028505fbe4349b51ec6748636ab45d411558368ffbd7dda439d27829d074402a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ada9e21dd1a7e88d4954002d67a65210f3ffd6aa0c74061e558aa46d9d06981e", "generation_seconds_batch": 17.85402303491719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9114010334014893, "target_z": 0.302734419760103, "ce": 6.252951145172119, "genuineness": 4.65797233581543, "identity": 0.6316850781440735, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/028505fbe4349b51ec6748636ab45d411558368ffbd7dda439d27829d074402a.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "029da403728a796d3490a571f9e26ad6824f2e679813fa7a10b3c8888fa6ee2b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7cbe6cdb8574bd8005c4e9e5e989a8e9e6dab4bc702c95b3e4b574c5cbfd0301", "generation_seconds_batch": 23.112488563987426, "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": 6.141753673553467, "genuineness": 4.20609712600708, "identity": 0.43788284063339233, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/029da403728a796d3490a571f9e26ad6824f2e679813fa7a10b3c8888fa6ee2b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "02d2ca1c1fc2183f36cafcd938bfaceba7160140b126e167d9d2271129fd50a6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9d9113bb956fbd4244e22124d2272a5c68dd351a08ae66898db78eb635d7dbf1", "generation_seconds_batch": 23.94466196396388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.806497097015381, "target_z": 3.7812498567676704, "ce": 5.804190158843994, "genuineness": 3.7165660858154297, "identity": 0.2519351541996002, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/02d2ca1c1fc2183f36cafcd938bfaceba7160140b126e167d9d2271129fd50a6.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0309fbedf174c1c13bd74e0dbd00fb5fead0d27ede7cda261dc0176c0917586c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0ce11e6586e54cbf7fe0d8ac416584a62e46b3d8b66b7817ab03c6318dcaebd5", "generation_seconds_batch": 24.813554077991284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4020261764526367, "target_z": 2.0937501432323296, "ce": 6.157639026641846, "genuineness": 4.030367851257324, "identity": 0.6854501962661743, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0309fbedf174c1c13bd74e0dbd00fb5fead0d27ede7cda261dc0176c0917586c.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "0328e4a28552714a322175185181b0c068eef7f6a9b790523a4d2cfb76e4897f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "9b866b2a677191bc8ac9c5dc5e7d6a4fed0e0eebd5af4bfac27ad501c53064ef", "generation_seconds_batch": 16.93509841291234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8428385257720947, "target_z": 1.4218750716161648, "ce": 5.383228778839111, "genuineness": 4.356721878051758, "identity": 0.7577547430992126, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0328e4a28552714a322175185181b0c068eef7f6a9b790523a4d2cfb76e4897f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "03782c0bf125b944d281747a0261084a24b9430da5a504fe52db40772901ee98", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "aa7986f90aaf3f9635ccb0a42f993392e60fa848344143b6cac081acad1b6346", "generation_seconds_batch": 21.835536123020574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15627908706665, "target_z": 3.0, "ce": 5.319687366485596, "genuineness": 3.9927117824554443, "identity": 0.415425181388855, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/03782c0bf125b944d281747a0261084a24b9430da5a504fe52db40772901ee98.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "03abc257177a1dbcd1c1c4d1adedd223ee855fca8163c5c7ee8eeffba2d544c1", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a117886c57dec802b461be41e61eab5ed5e3b26ae418cf1d42737d63e22e77a4", "generation_seconds_batch": 18.11708835104946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7403123378753662, "target_z": 0.09716795308396393, "ce": 5.95112943649292, "genuineness": 4.406930446624756, "identity": 0.7505401372909546, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/03abc257177a1dbcd1c1c4d1adedd223ee855fca8163c5c7ee8eeffba2d544c1.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "03e4a85c70974cf7bc377915d03b023dc791df79e739fea70aaa9688030ed196", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7dc103067f74edd7b39de5634d24fc766b6b9a607499c1a15c45f7bb58acea56", "generation_seconds_batch": 23.64414112304803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537235975265503, "target_z": 1.0546875358080825, "ce": 5.816103935241699, "genuineness": 4.293961524963379, "identity": 0.4419533312320709, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfuhren.", "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_high/03e4a85c70974cf7bc377915d03b023dc791df79e739fea70aaa9688030ed196.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "043c07e594321e124bdeeaa01bf86c7d9f057f3b03860e9ad08cb7ea734c884d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "637db35afe53e8155ee22ea22c8c472f0ae7814184ac4fb73b3f459284d609c3", "generation_seconds_batch": 22.922349330969155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6756969690322876, "target_z": 0.01953130371212364, "ce": 6.205295085906982, "genuineness": 4.770940780639648, "identity": 0.7228291630744934, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/043c07e594321e124bdeeaa01bf86c7d9f057f3b03860e9ad08cb7ea734c884d.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "046e04b54ff2c0024a1cbe89157f4332024c8b3194f6d4f9844d3cd03a58b773", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6a3fd8e4b51b4eb54bc573b1e0b3a191088876583b699aafdafacf7e8a6df88e", "generation_seconds_batch": 23.713894524029456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.310995578765869, "target_z": 1.9843750716161648, "ce": 6.292664527893066, "genuineness": 4.193545341491699, "identity": 0.3001599907875061, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/046e04b54ff2c0024a1cbe89157f4332024c8b3194f6d4f9844d3cd03a58b773.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "049b28bffec47dacb07e037cbb4aacd5d0d28665f679eba06deb6b305f385edb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8177172fa62115299607547c8f988a312aa44ebbbc2f22a0e6b48425d119c420", "generation_seconds_batch": 21.25012888503261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.948209285736084, "target_z": 2.75, "ce": 5.784333229064941, "genuineness": 4.030367851257324, "identity": 0.6397514343261719, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/049b28bffec47dacb07e037cbb4aacd5d0d28665f679eba06deb6b305f385edb.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "04a73a44c1dbc3a65e75d9afd8f20b515a5ae621786fb01807c6f35000f3e8d2", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "fdd25ea42efd63dbd8c5f22485757312822f89acafb46d0f8c2423b87518a30d", "generation_seconds_batch": 16.818530895048752, "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": 6.054384231567383, "genuineness": 4.262581825256348, "identity": 0.7285153865814209, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/04a73a44c1dbc3a65e75d9afd8f20b515a5ae621786fb01807c6f35000f3e8d2.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": "04fc687d294c3162f6037a9007237599c839f19bd0790e268f65624eaf85f4ba", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "956c1f7eff0a4ae46633e7330a1297a54c082d5b57f0d6aa6d60b50baa52a9ef", "generation_seconds_batch": 18.068316887016408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6751177310943604, "target_z": 2.421875071616165, "ce": 5.808161735534668, "genuineness": 4.250029563903809, "identity": 0.6222444176673889, "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_high/04fc687d294c3162f6037a9007237599c839f19bd0790e268f65624eaf85f4ba.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "05847776cf5d3862b2ed9c35ab5ce599d5af4970cec5146195e4c9b74e034b2a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6e2d0904e27d58b1198052201e65d6276f0a12ed292e699d0b2f9df118bc2b0b", "generation_seconds_batch": 22.676186230033636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.34867262840271, "target_z": 0.8281249283838351, "ce": 6.09409761428833, "genuineness": 4.68307638168335, "identity": 0.7303709983825684, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/05847776cf5d3862b2ed9c35ab5ce599d5af4970cec5146195e4c9b74e034b2a.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "f0eac0ccbbd3bd2c7410c059f8bddb00768684a4d568f5dba915ccac012b5960", "generation_seconds_batch": 15.693938454031013, "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": 6.348263263702393, "genuineness": 4.7332844734191895, "identity": 0.441270649433136, "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_high/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "05b91c71093ae1aaa0bea35d0d03cbc17101edbc369140daa5cead787bf22c2c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "86b7e35c77c54ab0e817c5cb8797625d602a85caa8f294715ad583f56ac00cd4", "generation_seconds_batch": 22.75012099801097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.68678617477417, "target_z": 1.2343750716161648, "ce": 5.935244083404541, "genuineness": 4.406930446624756, "identity": 0.5963869690895081, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/05b91c71093ae1aaa0bea35d0d03cbc17101edbc369140daa5cead787bf22c2c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "05f69111a79bf7ca5345aee3c7a8150655b84644ad4f504a80530475ffabe0f5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0eba79a428b6ae8a88728d00ec6bca2b7bfaa57b781c40b614fcc25421d7a165", "generation_seconds_batch": 17.4862761850236, "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": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.4622882306575775, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/05f69111a79bf7ca5345aee3c7a8150655b84644ad4f504a80530475ffabe0f5.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "063dfb395108488ea7b3ae1f268f0c7a773c1fdde4f46e5642ce9019d987d231", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f3a67c61f9307e3fd9bb9b6d2ada4c91595f6eebf4a67b662f0e7856fc1dc418", "generation_seconds_batch": 22.638067246996798, "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": 5.919358730316162, "genuineness": 4.7332844734191895, "identity": 0.5889293551445007, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/063dfb395108488ea7b3ae1f268f0c7a773c1fdde4f46e5642ce9019d987d231.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "06919a46c705282172ba3f70fa49923f3ff49fed431f75ae46ca3954f7f64485", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "885b2bf9590426ab91130ecd559f1acd8a40cab0564ddc945384ecfcbdfc75e6", "generation_seconds_batch": 17.28131251502782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 5.665193557739258, "genuineness": 3.9676074981689453, "identity": 0.4094584584236145, "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_high/06919a46c705282172ba3f70fa49923f3ff49fed431f75ae46ca3954f7f64485.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "06c9a30b3f40708b0e46e61b24370d42590df48e5746b89d930744b0c25a1a22", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "38efdbe1abcd2fc6342951ab63397f54967754eb091dce9082da0b13247aaf90", "generation_seconds_batch": 17.59279162599705, "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": 6.515059471130371, "genuineness": 4.337893962860107, "identity": 0.6303917169570923, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/06c9a30b3f40708b0e46e61b24370d42590df48e5746b89d930744b0c25a1a22.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "06e4dc67a5afa7721d04c4af408ed4a143dd1072cb77ede00dc3d8ae30a9eefc", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d535f834ff33b93c5b25c046ae56891dc2c0f132b81ab82d3d1939ea480f5385", "generation_seconds_batch": 17.177780136931688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2641444206237793, "target_z": 0.7265626074242473, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.6551125645637512, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/06e4dc67a5afa7721d04c4af408ed4a143dd1072cb77ede00dc3d8ae30a9eefc.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "074a183a4706d100ea134b18d104d11043793371ca4acdf337b7fa255a75822a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f52c3a8998621423b38e72ebac34b12c9a43fc507ffd540eff1f3362d1a0da9", "generation_seconds_batch": 29.32788363506552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.312331199645996, "target_z": 3.1874997135353405, "ce": 5.6254801750183105, "genuineness": 4.118232727050781, "identity": 0.7414376139640808, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/074a183a4706d100ea134b18d104d11043793371ca4acdf337b7fa255a75822a.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0767b570bf772319eb9e2005125c2837a1013df7b888e95ad4238bdda5d2cf55", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bf3d1120be37555bb56caad9f2c8183c5c173001a39df606f83a71e9d2c4a581", "generation_seconds_batch": 23.93550693197176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 6.133810997009277, "genuineness": 4.042920112609863, "identity": 0.6294962763786316, "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_high/0767b570bf772319eb9e2005125c2837a1013df7b888e95ad4238bdda5d2cf55.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "078e8a72b5ad1633db74cc97c48cbe67bb399ee40f54cc6fd8050a1aefef84da", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a81e752575d22f75344bdf9b812a1a34119b926f0264bf57895b41e1a7998f88", "generation_seconds_batch": 23.82584017305635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.766148090362549, "target_z": 2.5312498567676704, "ce": 5.95112943649292, "genuineness": 4.042920112609863, "identity": 0.510590136051178, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/078e8a72b5ad1633db74cc97c48cbe67bb399ee40f54cc6fd8050a1aefef84da.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "079df5197a9db3986d2fddb59430d7a2a7588aabc7b85108ceb0ea7ef59d8be2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "dd85d5513dcae1c9c5dc6687dd7beb0ba7aa1eadc8304259a0f46ad807845817", "generation_seconds_batch": 24.205222241929732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9585418701171875, "target_z": 0.35937507161616483, "ce": 6.427689552307129, "genuineness": 4.319066047668457, "identity": 0.4403448700904846, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/079df5197a9db3986d2fddb59430d7a2a7588aabc7b85108ceb0ea7ef59d8be2.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "07c4a51315d92effaa00cf039803fe4f6c438321c418e661ef3be87a0c494da1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "63b9806a8c1d8b5aeea845027b2e2d4f1ff9e2a53d21fa6594c61f58338c4917", "generation_seconds_batch": 24.97288221505005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.722024917602539, "target_z": 0.07519525431186605, "ce": 6.086154937744141, "genuineness": 4.68307638168335, "identity": 0.7082734107971191, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/07c4a51315d92effaa00cf039803fe4f6c438321c418e661ef3be87a0c494da1.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "08a903142a9cc5973e9437290f391da2ae958974a1b79a821798c5a4907bfd26", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b1a0b865b9d847216efd2acc3e5ef39f6ec22eafc267722397d7bbf5c8768cd9", "generation_seconds_batch": 18.611502249957994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.491720676422119, "target_z": 1.0, "ce": 6.165581703186035, "genuineness": 4.494794845581055, "identity": 0.5895141363143921, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude graam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/08a903142a9cc5973e9437290f391da2ae958974a1b79a821798c5a4907bfd26.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "08c8abd8c4205e2f8d6c32c9c051ac251b1202cd20f3a24f7df64e43c027ebad", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ccfb1f1b1b28c68c33229428a6ce121ae0144a17b82c997a0b6fb63b8af897aa", "generation_seconds_batch": 22.725682756048627, "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": 5.879645347595215, "genuineness": 4.231201171875, "identity": 0.548773467540741, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/08c8abd8c4205e2f8d6c32c9c051ac251b1202cd20f3a24f7df64e43c027ebad.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "09457f242511d7b89288ca09d69ba8384ed776ceffc731a3d534d857d3dce080", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "a6ab9c4714340175da38711c7d70e28eaa694cab3dd39b9a4b36069ae81169d1", "generation_seconds_batch": 23.66000220598653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8181657791137695, "target_z": 2.5937501432323296, "ce": 5.744620323181152, "genuineness": 3.973883628845215, "identity": 0.5074067115783691, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/09457f242511d7b89288ca09d69ba8384ed776ceffc731a3d534d857d3dce080.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "0963d63373d11d5fc33d7debd3772f939d6c5e38dc85d16c35495a27f3dcbe90", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "738439a1d944392c62666b8d3376df725bf5b0fdaefc09583f54c0bad38eaf52", "generation_seconds_batch": 23.396226670010947, "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": 6.252951145172119, "genuineness": 4.519899368286133, "identity": 0.6982494592666626, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0963d63373d11d5fc33d7debd3772f939d6c5e38dc85d16c35495a27f3dcbe90.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "0970aa0826948107700043e2011544ac499e69fa34c6b6594f33ed1ac2ba061e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "44c3519645769b1446ac16e3e9e4a74561fa839b1f46f034cac129d63d9908f5", "generation_seconds_batch": 22.567111302050762, "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": 6.252951145172119, "genuineness": 4.607763767242432, "identity": 0.48812732100486755, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0970aa0826948107700043e2011544ac499e69fa34c6b6594f33ed1ac2ba061e.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": "0974ed18214abc9749d66bfc74d0a838469e942b2fe9fdc46d7a8f78a43d18f5", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c515fe9c309167c8ac3dceab274f0745f4a6bb11b2d9160e08782b125962842c", "generation_seconds_batch": 17.412617385969497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.01323127746582, "target_z": 2.8281252148484946, "ce": 6.205295085906982, "genuineness": 4.055472373962402, "identity": 0.2673245072364807, "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_high/0974ed18214abc9749d66bfc74d0a838469e942b2fe9fdc46d7a8f78a43d18f5.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "09c8794a13fe1fb37ddbeeb321d4b04b48ac2d088c0df7bd4bf0cf5a16804f48", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "478492d4e012a3619d54371c97241fe5a6337de2ad6ef8d2304d165905fb2e0c", "generation_seconds_batch": 17.151638868963346, "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": 6.205295085906982, "genuineness": 4.896461486816406, "identity": 0.6234971284866333, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/09c8794a13fe1fb37ddbeeb321d4b04b48ac2d088c0df7bd4bf0cf5a16804f48.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "09d208fcf6d8085878648a5af140fed9f58da9e984f7906c0de9c566fb076d51", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "15261f7196f5cc562ea54bbbc2586b90271046bbaa3769898b45f502d8ab6a5b", "generation_seconds_batch": 21.823137464001775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.286322593688965, "target_z": 3.1562498567676704, "ce": 5.403085231781006, "genuineness": 3.7165660858154297, "identity": 0.5129395127296448, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/09d208fcf6d8085878648a5af140fed9f58da9e984f7906c0de9c566fb076d51.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "09d740d1ca869cbb4b87e7eb5f38f729b66fc68f70bfcbc13f7ac5d480c75c8d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "28a23c4f40d0d674c778507b5ea736208c6e3b3291bfb38dd5e61625868192e7", "generation_seconds_batch": 17.959557310910895, "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": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.7587043046951294, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/09d740d1ca869cbb4b87e7eb5f38f729b66fc68f70bfcbc13f7ac5d480c75c8d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "0a61a59eefbb4d390a1c685f7114ef34a39cc3ac7ec9a8d3e48a16b466c3f76f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "abdb9d4787e7fad210a91d308f7685b05214a38646ccf08864bf6b88fa2d6272", "generation_seconds_batch": 22.622132421005517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4787163734436035, "target_z": 0.9843750716161649, "ce": 6.24500846862793, "genuineness": 4.331617832183838, "identity": 0.44588035345077515, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0a61a59eefbb4d390a1c685f7114ef34a39cc3ac7ec9a8d3e48a16b466c3f76f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "0aae6e179a642d5dab0c786c04ebf84793dea1e25c45c4497c219a0d891f0e48", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "78ed122f04f293e903071d7fe6ec9ad91e69fab787a5101a836bbaa6d3a3762d", "generation_seconds_batch": 25.02847490296699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9663798809051514, "target_z": 1.5703124641919175, "ce": 6.117925643920898, "genuineness": 4.419482707977295, "identity": 0.6416064500808716, "hypothesis": "La porte 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_high/0aae6e179a642d5dab0c786c04ebf84793dea1e25c45c4497c219a0d891f0e48.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "0ad82991393662b7007501477a913f11b125d2d5e61f2554b2c9d1fa69d4285e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3e4b141ba552311a58e363eaec48f44dba6e7e41b5b47d235afbda658d1ccff2", "generation_seconds_batch": 16.184966755099595, "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": 6.395919322967529, "genuineness": 4.369274139404297, "identity": 0.5854443311691284, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0ad82991393662b7007501477a913f11b125d2d5e61f2554b2c9d1fa69d4285e.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "0b95351dcbc00a9c7be12f5c3c1f9c1604347c5c76dff1606ce24040ba64e5f5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e9996da90f0ed256fdc760a52142f6a84c3fd2b8db042ace3a47a916ad4d24bd", "generation_seconds_batch": 23.962517819018103, "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": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.4903109073638916, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0b95351dcbc00a9c7be12f5c3c1f9c1604347c5c76dff1606ce24040ba64e5f5.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "0bb4f309c06e7e6dbc61b08bcb18ab1caff1ba0e97854e4cf46dfeae9e1fe3cf", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "783896155a090a567b02354c16f8d5b6c3fb14276a4ee81744b5974c95e9c13e", "generation_seconds_batch": 18.1122174190823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7791526317596436, "target_z": 2.546875071616165, "ce": 6.213237762451172, "genuineness": 4.005263805389404, "identity": 0.3492518663406372, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0bb4f309c06e7e6dbc61b08bcb18ab1caff1ba0e97854e4cf46dfeae9e1fe3cf.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "0c71e953897a7986d22d64103905a41ec70ccb173c65be1455615c69c77738cb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9df6e118230767de32aab74f84b80081ece84bdfdc2c36e1f7162eb37101cb7c", "generation_seconds_batch": 22.145988740958273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5840871334075928, "target_z": 2.3125, "ce": 5.319687366485596, "genuineness": 4.193545341491699, "identity": 0.6008149981498718, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0c71e953897a7986d22d64103905a41ec70ccb173c65be1455615c69c77738cb.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0c8f3fed1a08811696c492cb4aca35b4a8c3196518530c7539bdbb1085484a60", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fcde20cb18c977621cf5d62ffc6c628045758725fd2d09951a3bedd0dc72658b", "generation_seconds_batch": 16.063230513012968, "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": 5.637393951416016, "genuineness": 4.444586753845215, "identity": 0.8354383707046509, "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_high/0c8f3fed1a08811696c492cb4aca35b4a8c3196518530c7539bdbb1085484a60.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "e7e9c4f35ecc5997ea6ef93c1a401b450f083b44c4bbd520c7baa8b687ed2ae4", "generation_seconds_batch": 16.53020427795127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.054384231567383, "genuineness": 3.9801597595214844, "identity": 0.49156710505485535, "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_high/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "0d0bab216ccf541d4957db8516aa8d39196f19603b30d8c1b56eb6cd493fe512", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "986e8b05038c0b6c480bd541a08654f8549d7f0c71e9227555841c5cc53b44bd", "generation_seconds_batch": 22.32308379805181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.898396611213684, "target_z": 0.2871093481439382, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.6214817762374878, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0d0bab216ccf541d4957db8516aa8d39196f19603b30d8c1b56eb6cd493fe512.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "0d313205820429d061942cd5df036cfee9ed1112b075fd9c8391d13b33cc12c6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "feab05a171a574dd104e5dab702a8ebf026e379d08eb9012464041735cb78669", "generation_seconds_batch": 17.54144533595536, "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": 5.7128496170043945, "genuineness": 4.369274139404297, "identity": 0.5537630319595337, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer.", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0d313205820429d061942cd5df036cfee9ed1112b075fd9c8391d13b33cc12c6.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0d8c47b662523fca06c45c0b0c6d9db1daea9001104ff8a485a2ed29e0884a82", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8201bab42538d7b7f86a4ceb22196444643c6c96a49712b56bebe10ea027653d", "generation_seconds_batch": 16.406589817954227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5663789510726929, "target_z": -0.11181646667613909, "ce": 5.990842819213867, "genuineness": 4.268857479095459, "identity": 0.6184839010238647, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0d8c47b662523fca06c45c0b0c6d9db1daea9001104ff8a485a2ed29e0884a82.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "0d9a4e48e2ebcb97d500c0c19430a0223aa805db3fd625d37b1405c8af3289b0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a543fb4a4ffefe533b1f1518181324d3c4b816fca306860e84cc0d225232a0a2", "generation_seconds_batch": 23.260423870990053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823331832885742, "target_z": 1.3984373925757527, "ce": 5.982900142669678, "genuineness": 4.306513786315918, "identity": 0.4875154495239258, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0d9a4e48e2ebcb97d500c0c19430a0223aa805db3fd625d37b1405c8af3289b0.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0dc6865918d7f0f3cf0fe5c84092a648e60f7c4df3d2a47b270d3cc326086b60", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e536e7b5b5f7faa0b1b5345c5ec33276a18980a7cb1697f5110e6cc1fcc1b5f3", "generation_seconds_batch": 21.04347609798424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 5.895530700683594, "genuineness": 4.344170093536377, "identity": 0.5026895403862, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0dc6865918d7f0f3cf0fe5c84092a648e60f7c4df3d2a47b270d3cc326086b60.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0de7e6d1aa855d9d72f454c9a133f0b3230d135e27f89f9afdac6e3208a5c006", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4d0154d5ccf6afccedd09e98bff127bffd7cc294b14550ed49f56b97a1278b5c", "generation_seconds_batch": 16.32634205603972, "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": 6.0146708488464355, "genuineness": 4.344170093536377, "identity": 0.46490222215652466, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0de7e6d1aa855d9d72f454c9a133f0b3230d135e27f89f9afdac6e3208a5c006.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "0e0dda37045b3077f2d77fe4a1d2acab54090a0a0d964004fa45b661a3458a63", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3daa1004fddfeac17b7cf38525ddfeeec8b9c826c1b0a574dacaf22d1a142de4", "generation_seconds_batch": 23.093001386034302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1939563751220703, "target_z": 1.8437501432323298, "ce": 6.149696350097656, "genuineness": 3.9676074981689453, "identity": 0.6888865828514099, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0e0dda37045b3077f2d77fe4a1d2acab54090a0a0d964004fa45b661a3458a63.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": "0e26433442cecd4c2f68850afd2d6855f267d0eb6a0e12fcc2adba99e429aa29", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70af1c64834a4070ac71a963c7ce51b00602aa0508671d5e6ca0b3d8c0bca753", "generation_seconds_batch": 24.911018306971528, "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": 6.022613525390625, "genuineness": 4.356721878051758, "identity": 0.7409835457801819, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai eux gardées dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0e26433442cecd4c2f68850afd2d6855f267d0eb6a0e12fcc2adba99e429aa29.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "0e38b1f5285f67a398443ec72066c3fe1c50b7fad79ad6b7b1da57193f80f626", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d2f52b624c29542f4ce07499d18212332907dc323c14b32dda22e83938b13c3f", "generation_seconds_batch": 21.25012888503261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1744496822357178, "target_z": 1.8203124641919175, "ce": 5.7128496170043945, "genuineness": 4.293961524963379, "identity": 0.5227729678153992, "hypothesis": "He encontrado la cajita. Estaba junto al. junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0e38b1f5285f67a398443ec72066c3fe1c50b7fad79ad6b7b1da57193f80f626.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "0e579cd5671167b2aa3929cde379b7518d45e8ee11a4fd004e0e203a1576eea1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a56337fa49e5603c5aee11f140399bc2853fceac9481f9bf7ba17b028437485e", "generation_seconds_batch": 16.73251497000456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4150304794311523, "target_z": 2.109375071616165, "ce": 6.006728172302246, "genuineness": 4.168440818786621, "identity": 0.6152697801589966, "hypothesis": "De deur 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_high/0e579cd5671167b2aa3929cde379b7518d45e8ee11a4fd004e0e203a1576eea1.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "0e764d217457318e6422aada056188ea5eda1b46b3eca8c2e9aacdb4b197f75a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "63ad1efbf9efbf484c4914e56a707e04b090ef8d8bdbbf06e45294df896be052", "generation_seconds_batch": 16.95107209100388, "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": 6.332377910614014, "genuineness": 4.174716949462891, "identity": 0.30982357263565063, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0e764d217457318e6422aada056188ea5eda1b46b3eca8c2e9aacdb4b197f75a.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": "0e920d830a96141c14255c38694da7c7daf76c5ed3c6846f90a2959caca1450c", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29e4130b3fe62b988624cede2fac876a3a4d8c3005b4450b1862cfda31394897", "generation_seconds_batch": 16.316368971951306, "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": 5.967014789581299, "genuineness": 4.3504462242126465, "identity": 0.6270785331726074, "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_high/0e920d830a96141c14255c38694da7c7daf76c5ed3c6846f90a2959caca1450c.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "0ea6f759d1d552e585ff7fcd3f1fec6c9d9aad27777220455e0eb176bfff0d4b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fd45a2e7c2905aa8b73c2a21b512ace95f1419203c981c9152c83af9f3aa331a", "generation_seconds_batch": 22.145988740958273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.389021635055542, "target_z": 2.078124928383835, "ce": 5.244231700897217, "genuineness": 4.036643981933594, "identity": 0.468328058719635, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0ea6f759d1d552e585ff7fcd3f1fec6c9d9aad27777220455e0eb176bfff0d4b.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "0ec678772bf282027cc91c9f358360e6ddbe4a05a5bedf96ecc56904fa3720aa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2a8e5f66903f70e42149f303f8b8af7a445cfeb7bc480169ec752ea5f1013d9f", "generation_seconds_batch": 23.493406652007252, "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": 6.157639026641846, "genuineness": 4.18099308013916, "identity": 0.3891884684562683, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0ec678772bf282027cc91c9f358360e6ddbe4a05a5bedf96ecc56904fa3720aa.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "0f473d94f64a07ef6da1fb13d70eb935fd28247328de9247bc749d301508472f", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "617fbff1ad3e240e45c301d162303c8b6da118c55f49be4751fb56430789f970", "generation_seconds_batch": 17.688341216999106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.232969284057617, "target_z": 1.8906249283838352, "ce": 6.0146708488464355, "genuineness": 4.093128204345703, "identity": 0.4666084945201874, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/0f473d94f64a07ef6da1fb13d70eb935fd28247328de9247bc749d301508472f.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "0fde476118233410aaf7c28c8e32d784d410e0caff9a20781d93fb3cdedeee37", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "871892413822b2cfe3a7aa8eb8ea4918b7c4dbcce68e1a6c0a38175756e04799", "generation_seconds_batch": 16.406589817954227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0723299980163574, "target_z": 0.4960937679040412, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.6742756962776184, "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_high/0fde476118233410aaf7c28c8e32d784d410e0caff9a20781d93fb3cdedeee37.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": "1005f0a1f1a6057604d38910c0424ec26f9dee8486cc347cbc3a9110b9071ac8", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "13e7ea3f87b44d75aa7fcffb6b42f8d93364050d76695148bb824d78ea61a368", "generation_seconds_batch": 23.850330410990864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.870182991027832, "target_z": 2.6562498567676704, "ce": 5.895530700683594, "genuineness": 3.9927117824554443, "identity": 0.23109714686870575, "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_high/1005f0a1f1a6057604d38910c0424ec26f9dee8486cc347cbc3a9110b9071ac8.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "102d238ffae800ec1315b8e9cf5ffa1ce7e838cd39d426c4683591230731e35d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8fa20d70ec95a9007fe58d3a687aacf1e793090405a3fa6d8580592f628b0cb0", "generation_seconds_batch": 18.59255115792621, "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": 6.340320587158203, "genuineness": 4.482243061065674, "identity": 0.3988565504550934, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/102d238ffae800ec1315b8e9cf5ffa1ce7e838cd39d426c4683591230731e35d.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "107964be116f70eb576106844a5fd1361c82bd190b2782ce5ad614844b958cc9", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ce89c792646605d85c8a2a06872bba02471d861f9691d3161815dff6e14961ac", "generation_seconds_batch": 17.303496812004596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7354357242584229, "target_z": 0.09130860494002575, "ce": 6.2767791748046875, "genuineness": 4.808597087860107, "identity": 0.5859363079071045, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/107964be116f70eb576106844a5fd1361c82bd190b2782ce5ad614844b958cc9.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "107f098c165dd985faee45ab551e8ed0b0902b7bd9d4c31a06293b47b78c5426", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "113d4335205037ae83e98182934e095b6a8e928d2ab615ece84042c84f49e90a", "generation_seconds_batch": 15.75975651293993, "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": 6.141753673553467, "genuineness": 3.9550557136535645, "identity": 0.3108985722064972, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/107f098c165dd985faee45ab551e8ed0b0902b7bd9d4c31a06293b47b78c5426.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "109685db45edf488f417ea12f117f3e31f1f3a77443d586c83b2c8cc842a60c2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c6bf8c3681beb3927d55e12583f82c98b00e35d6693fa274a325bafc5f9d75ad", "generation_seconds_batch": 25.742135660955682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.078252792358398, "target_z": 2.9062498567676704, "ce": 5.685050010681152, "genuineness": 3.798154592514038, "identity": 0.4497631788253784, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/109685db45edf488f417ea12f117f3e31f1f3a77443d586c83b2c8cc842a60c2.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "10c1e27cb46c65e554a07afdfb5e6e19aa785a8ee0870a15aa70a1dde4d6e8c5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f484f112b80007cb11dbbb1d58b9cf920de0849cf985bd9ed59d45bd10d4b49f", "generation_seconds_batch": 15.746108162915334, "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": 6.284721851348877, "genuineness": 4.356721878051758, "identity": 0.5679982900619507, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/10c1e27cb46c65e554a07afdfb5e6e19aa785a8ee0870a15aa70a1dde4d6e8c5.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "10ff60cf0cd4f7410955cbe14fdad715e841099028cc27631f0b756b373f2b2e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b5b46767630e16075498ba768f1d872f037f58b5efc5fc8dfdf34376bc5fb62c", "generation_seconds_batch": 16.595890511991456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.16794753074646, "target_z": 1.8125, "ce": 5.677107334136963, "genuineness": 3.9801597595214844, "identity": 0.3461654484272003, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/10ff60cf0cd4f7410955cbe14fdad715e841099028cc27631f0b756b373f2b2e.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "1134b4df433e2886e17955eaeeeae5eabf729993b70fa117fadcbe87e70d1522", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "61ffdae53eda9e2523e19ee03056ce4e667cbeb4b48e9f01b710d766dbee1642", "generation_seconds_batch": 22.13491012295708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7388036251068115, "target_z": 1.2968750716161648, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.5279317498207092, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1134b4df433e2886e17955eaeeeae5eabf729993b70fa117fadcbe87e70d1522.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "11679b54f2ff1cd84b8c5411db5ea5c6028fe213a12df9cf8525a1c075c067f3", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ddeea1ef7478738cb5a90cddda4ad5c5d56dac57088c986068246a99d4791944", "generation_seconds_batch": 17.98971182096284, "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": 6.181467056274414, "genuineness": 4.457139015197754, "identity": 0.8199979066848755, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/11679b54f2ff1cd84b8c5411db5ea5c6028fe213a12df9cf8525a1c075c067f3.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "1175751eaab2f4065052917f3eb980ca7738fec0a08e6f7f0c01f9efffd236f6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "598b74d9f966e2142370fb7c072182d9771383974a3949e5aa7e47b002d7bcf4", "generation_seconds_batch": 17.17835149099119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6231002807617188, "target_z": 2.359375071616165, "ce": 5.816103935241699, "genuineness": 4.457139015197754, "identity": 0.5700207948684692, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1175751eaab2f4065052917f3eb980ca7738fec0a08e6f7f0c01f9efffd236f6.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "11bf0976cd469c29f219c5a27f3ee1cc423c0f891aff9c3d435d2feac209abb2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4de744cf5c5c867a1fb5a69fa0f765fcdc1bf10de609a0c0bdbf2c8a3a55c04d", "generation_seconds_batch": 23.99544641096145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.858514785766602, "target_z": 3.8437501432323296, "ce": 5.550024509429932, "genuineness": 3.785602569580078, "identity": 0.42122355103492737, "hypothesis": "Ik 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_high/11bf0976cd469c29f219c5a27f3ee1cc423c0f891aff9c3d435d2feac209abb2.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "11c29d52e8a796f62eab394b29409c288491bcd865d3fbb6b6cd902dd84eea6b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6e3c2b7197f6796287855ecaff66803e92b4fe1bbccd6470e207504f52f03610", "generation_seconds_batch": 22.813404069980606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.507396697998047, "target_z": 3.4218747851515054, "ce": 6.205295085906982, "genuineness": 4.011539936065674, "identity": 0.5137737989425659, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/11c29d52e8a796f62eab394b29409c288491bcd865d3fbb6b6cd902dd84eea6b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "1211bf74a0d9caddc2edb02bddde95a2efe8169268e758405ad72ac91b0e6b2a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "efdad485aa8ee5e8b3858e170f13fb183caea30af1e53a664c21891ef072a923", "generation_seconds_batch": 16.614855902036652, "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": 6.24500846862793, "genuineness": 4.557555198669434, "identity": 0.5626212358474731, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1211bf74a0d9caddc2edb02bddde95a2efe8169268e758405ad72ac91b0e6b2a.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "121a33fa9bc55b0c7a0874bd95365e8c137920b28ee8d3acbf8363758b13297e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b902ec3dcf6f65962504d27c724f1a785022a5478c3352f75b800b4241ff99e8", "generation_seconds_batch": 21.04347609798424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.454043388366699, "target_z": 2.1562498567676704, "ce": 5.895530700683594, "genuineness": 4.093128204345703, "identity": 0.43639203906059265, "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_high/121a33fa9bc55b0c7a0874bd95365e8c137920b28ee8d3acbf8363758b13297e.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "123a391f8a972245af32c6821c4b706c072058efd2cb9849991d916ab3e29e1a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "59a9643ed3116084eae7c84e3866eaa00822732af507bc1f2e1613c2b8d70e8c", "generation_seconds_batch": 22.622132421005517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.070269584655762, "genuineness": 4.20609712600708, "identity": 0.5356351137161255, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/123a391f8a972245af32c6821c4b706c072058efd2cb9849991d916ab3e29e1a.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "1285e33ca23512456a0ce2919dfc70f496c7996973d8fc64449870b9b74d8b23", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f064808a4dabd2cc5d6d72c002436f42aaccdf7202d7536553de1a3400321098", "generation_seconds_batch": 23.781660664943047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.611487865447998, "target_z": -0.05761720988005152, "ce": 6.038498878479004, "genuineness": 4.394378662109375, "identity": 0.6314423084259033, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1285e33ca23512456a0ce2919dfc70f496c7996973d8fc64449870b9b74d8b23.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "f2d8f76782a02b13485b9d2c3c7046c50bbae0407ecc29554adaae727619b89e", "generation_seconds_batch": 16.76059895195067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.491720676422119, "target_z": 1.0, "ce": 6.24500846862793, "genuineness": 4.444586753845215, "identity": 0.6152774691581726, "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_high/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "12e8fc0eb8c56dc38a52169e1d1e041ccb2bc7fc0c1b9716ff95abf4586d6693", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "235b0f52ae4797740f84e881a37f6d80f668df2aa691c036d57be9bd76b1c7c0", "generation_seconds_batch": 17.214929033070803, "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": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.49940311908721924, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in Keuken.", "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_high/12e8fc0eb8c56dc38a52169e1d1e041ccb2bc7fc0c1b9716ff95abf4586d6693.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "134dae5c6df4107c8f4373427865957e14d79cce36e421e10359512e7923dcd1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "09bee72d7549426fde5531f851cb51dc15a3e0a08b501c4c432488bfe242abf3", "generation_seconds_batch": 18.226447381079197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4163665771484375, "target_z": 3.3125002864646595, "ce": 6.1894097328186035, "genuineness": 3.8420867919921875, "identity": 0.42841455340385437, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/134dae5c6df4107c8f4373427865957e14d79cce36e421e10359512e7923dcd1.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "135f8fa4f744a2759b0407777ebc8b2ee8e38c0cdcfa549cda8b6cadaf9e3f20", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "caeb2c45dea85fe91fb16c139c7826c4beb6bd362aa18e8b54eeadb5ffab8332", "generation_seconds_batch": 16.751722809043713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9227797985076904, "target_z": 0.3164062320959588, "ce": 6.205295085906982, "genuineness": 3.8671910762786865, "identity": 0.2984923720359802, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/135f8fa4f744a2759b0407777ebc8b2ee8e38c0cdcfa549cda8b6cadaf9e3f20.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "139eb1b2d9e7eff9841e788d2e457950ca702cf030346788b25c8737748de9a6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f14a6c68dbaa5805f6435308822a686eb3fdacc62fefaa01d494d9af31182046", "generation_seconds_batch": 18.739263223018497, "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": 6.197352409362793, "genuineness": 4.394378662109375, "identity": 0.49498629570007324, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/139eb1b2d9e7eff9841e788d2e457950ca702cf030346788b25c8737748de9a6.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "13c72042b8d60b4f06b491202ebebd3074486cd1165f2f3ad2f8082953c3a253", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b0bb5605e685773b5c9a4808c29f0b7e91a51e3d6f52ffa2be5a1d2b4c9177d0", "generation_seconds_batch": 21.508463278063573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.608759880065918, "target_z": 1.1406249283838352, "ce": 6.252951145172119, "genuineness": 4.024092197418213, "identity": 0.39102649688720703, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/13c72042b8d60b4f06b491202ebebd3074486cd1165f2f3ad2f8082953c3a253.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "13e8754263873233b29f81f065dbd22169b30e36a32be3b8929ca235d0650ae8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "de560cef8dd7a25a023da36d50261e6533a4889a87a8c7ec8b6a942d1535880a", "generation_seconds_batch": 18.612191164982505, "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": 6.173524379730225, "genuineness": 4.7834930419921875, "identity": 0.6222025156021118, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/13e8754263873233b29f81f065dbd22169b30e36a32be3b8929ca235d0650ae8.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "14b4cedc1629c6c5672c3e67cb35b17aa887d8d6feb51168e6998825440e0186", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f2395c6194e0a24a0302f4b440c5c3134734a9658efc4163d3b226b4def48ed6", "generation_seconds_batch": 22.97565673198551, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.563127875328064, "target_z": -0.11572269877209788, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.680630624294281, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/14b4cedc1629c6c5672c3e67cb35b17aa887d8d6feb51168e6998825440e0186.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "14e8e04a9187632b5247574328e352ee8119efca7f22c6cd38df3f1732a71b83", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2898aa8df834051ef40ffe182707d4d54688a3aeddecae2e85c52f48a30b0bb2", "generation_seconds_batch": 17.283323684940115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.712794780731201, "target_z": 1.2656249283838352, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.5083618760108948, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/14e8e04a9187632b5247574328e352ee8119efca7f22c6cd38df3f1732a71b83.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "14ff619fec6cb7e6e335fd062a449f8ddb156175fa955f47c04ffd83e79c5f9a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c287e13c09615d27dad567135ce0e72fd697b2c68b794d39bb1cc9da03c3ab8b", "generation_seconds_batch": 17.28131251502782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.267395496368408, "target_z": 0.730468839520206, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.5270336270332336, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/14ff619fec6cb7e6e335fd062a449f8ddb156175fa955f47c04ffd83e79c5f9a.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": "150832debb4ca796aa56511a99515657668ec044fd34acb0076bbb234ae609b4", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5d27f7fb81871f0f3fcc6ca5c254755ea2e6fd4134c45729f105a8928134f53", "generation_seconds_batch": 17.236942888004705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 6.229123115539551, "genuineness": 4.482243061065674, "identity": 0.6623228788375854, "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_high/150832debb4ca796aa56511a99515657668ec044fd34acb0076bbb234ae609b4.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "151171a479e0af95f5b862934d8db9920fefd8ee41d45f07f9c2f3a611a822ef", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a3783475351a5bcff59c82ccfd84f1ad6c183b7f46af3826b026076b4823f2b4", "generation_seconds_batch": 23.252391926944256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.719296932220459, "target_z": 1.2734373925757527, "ce": 5.808161735534668, "genuineness": 4.6203155517578125, "identity": 0.6718270182609558, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/151171a479e0af95f5b862934d8db9920fefd8ee41d45f07f9c2f3a611a822ef.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "61fe6a14b02a2385a654cd4e6131bda9dc6ce1c42bcfa76e9aa9c69f6a0e7cdd", "generation_seconds_batch": 16.118621718953364, "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": 6.308549880981445, "genuineness": 4.645420074462891, "identity": 0.5079652070999146, "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_high/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "15a990be0bdc03d4c0fa5f5eff8eacff7139d96ab42f44ba308c665f92d622ab", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "59afb88adb13e500a57cf611048af5df25f4e94c944473cacb3f764a4d9c2132", "generation_seconds_batch": 23.44189335696865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.875349521636963, "target_z": 1.460937679040412, "ce": 6.086154937744141, "genuineness": 4.2374773025512695, "identity": 0.6699256300926208, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/15a990be0bdc03d4c0fa5f5eff8eacff7139d96ab42f44ba308c665f92d622ab.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "15e3e142e72272808cd8422a8f45b43753909e40db6d0505082b18b6ec7e8fe5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3f5e68e39ed1146cf2384077b362461e118ddb2e9d493b1258910e2700732c79", "generation_seconds_batch": 15.453905324917287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.758310079574585, "target_z": 1.3203124641919175, "ce": 6.133810997009277, "genuineness": 4.21237325668335, "identity": 0.6383754014968872, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/15e3e142e72272808cd8422a8f45b43753909e40db6d0505082b18b6ec7e8fe5.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "1656c9dba7f81cbf27a63ce6cb15630c1b76bbbd86d798234b072c0042f52b62", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e310185f1793cb5cccd90d6a050d7285a772212cba9883de9bad857a71a81fd1", "generation_seconds_batch": 24.54003436199855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8558428287506104, "target_z": 1.4375, "ce": 6.09409761428833, "genuineness": 4.118232727050781, "identity": 0.49355822801589966, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1656c9dba7f81cbf27a63ce6cb15630c1b76bbbd86d798234b072c0042f52b62.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": "16bf9378f047114218b5a7f9be269f26beb447ca03e42ec8425675181417af89", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f15b4aa39ee24f3a6bd9650bd18ad92003b223071e7bb023d24630a16dc71422", "generation_seconds_batch": 18.848783902009018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.206960678100586, "target_z": 1.8593750716161648, "ce": 6.038498878479004, "genuineness": 4.193545341491699, "identity": 0.39224299788475037, "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_high/16bf9378f047114218b5a7f9be269f26beb447ca03e42ec8425675181417af89.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "1710135083f5990069b2ad73320aceb7eae5fe6d412a38454d003fd7bcf6cd65", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8d377d90d3c5e67098e5c65683b980adc6acfb40b86007e14e76e28a3c7c07fc", "generation_seconds_batch": 17.639247552026063, "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": 6.0146708488464355, "genuineness": 4.7834930419921875, "identity": 0.5318247079849243, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1710135083f5990069b2ad73320aceb7eae5fe6d412a38454d003fd7bcf6cd65.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "172a3a94d959c152332cac3f2a3bff105b177c1305a4183f6c670b8c3d7ad4d5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "209b38c012e585117bee7cd01f701793aea045f077b0e333e20f8aa0278ebf75", "generation_seconds_batch": 24.012570268008858, "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": 6.070269584655762, "genuineness": 4.833701133728027, "identity": 0.675480842590332, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/172a3a94d959c152332cac3f2a3bff105b177c1305a4183f6c670b8c3d7ad4d5.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": "17724af9da155360ec6a08bfc4e125cbcbb086d8b6f9564ada41917c12bb6fb4", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96f7e16f7d0bf4749c65b6be7604e2c8b61f823beec472f799cf400f4a5faadf", "generation_seconds_batch": 22.045719899004325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.024899482727051, "target_z": 1.6406249283838352, "ce": 5.895530700683594, "genuineness": 4.381826400756836, "identity": 0.649010419845581, "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_high/17724af9da155360ec6a08bfc4e125cbcbb086d8b6f9564ada41917c12bb6fb4.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "17bbf8b67c82ee010e668aae54a53de98a31fc653585d7999f3a02e2314bb846", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2290dc461df0bac4187a39a18c211e262162029269145a480a895eafc1bbacfb", "generation_seconds_batch": 16.628088337020017, "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": 4.519899368286133, "identity": 0.6781860589981079, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/17bbf8b67c82ee010e668aae54a53de98a31fc653585d7999f3a02e2314bb846.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "18020b50dbaca743f70e89a080f0314ed9bd21cf8987cb77f83f2fbeb614b4bc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "76302540748bf799abb986b0d58bbff411aae6fc24f3fc694f95a0bda25e25f8", "generation_seconds_batch": 21.5504004759714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.089921474456787, "target_z": 1.7187501432323298, "ce": 5.4825119972229, "genuineness": 4.105680465698242, "identity": 0.8620861768722534, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/18020b50dbaca743f70e89a080f0314ed9bd21cf8987cb77f83f2fbeb614b4bc.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "180f099abe8230a35446c8af2630b31ddad284098357bb8d28f7cd0164c87d85", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "edeb4936b3969cf5e0a6542a0f6b555176ea4b5321bf46c067a6e873bd47e03b", "generation_seconds_batch": 21.553370086941868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2608933448791504, "target_z": 0.7226563753282885, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.7408517599105835, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/180f099abe8230a35446c8af2630b31ddad284098357bb8d28f7cd0164c87d85.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "180f1059630421143c3d7f39cf41007af817f74f2ddbe128d34330dc81eaff05", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1d852e1a3b2316ce118565f8e6b1b88f6d0b331b31a8b56443d59fa9db91c2d4", "generation_seconds_batch": 17.9918425999349, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.933868885040283, "target_z": 1.5312498567676702, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.753211498260498, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/180f1059630421143c3d7f39cf41007af817f74f2ddbe128d34330dc81eaff05.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "188a7bdcac07788c282c8a007e4ea703f7dad4625a7c3adddd869c4099bf07f4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f0998a364e71dda41da39cf1d2241a999d484ab190118fb991120968e1464ed9", "generation_seconds_batch": 23.673339955043048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.641270875930786, "target_z": 1.1796875358080825, "ce": 6.141753673553467, "genuineness": 4.319066047668457, "identity": 0.43106478452682495, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/188a7bdcac07788c282c8a007e4ea703f7dad4625a7c3adddd869c4099bf07f4.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": "18b5a0bef9c8fe157cd0b75ae1ff818f84cffa69d29dd9de2f69bb66d19a959b", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94565eb719202b5de36864786c90de433ef6c92c38c0d10f0be38346f9a2400a", "generation_seconds_batch": 23.312467143987305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.31941294670105, "target_z": 0.792968839520206, "ce": 5.6532793045043945, "genuineness": 4.413206577301025, "identity": 0.8889138698577881, "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_high/18b5a0bef9c8fe157cd0b75ae1ff818f84cffa69d29dd9de2f69bb66d19a959b.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "18c4accf54a922afc77431e9703da8002b1752029388d011260e8674fc53af48", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "96f1c44a4142aee6e7eacd695608d7480217e136269343dee7f3da5ce5444ad2", "generation_seconds_batch": 22.414218489080667, "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": 4.381826400756836, "identity": 0.6651488542556763, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/18c4accf54a922afc77431e9703da8002b1752029388d011260e8674fc53af48.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "18f22049db49cc6231b82706bfbf70975d828980d15a0c2d02bd26e43272e970", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "16a27744777726e1b61a10f0bbc6b165abb05d6803bd5ff14423dfc852e335cc", "generation_seconds_batch": 16.630346002988517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7581933736801147, "target_z": 0.11865237284406698, "ce": 6.41180419921875, "genuineness": 4.243753433227539, "identity": 0.5852380990982056, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/18f22049db49cc6231b82706bfbf70975d828980d15a0c2d02bd26e43272e970.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "199455d88ecdeb47f58dfb2a57417781929c4838cc517877e58db0ee7371470f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "95f270f6c204f38925d677303ca7b868088f9ae20eeb97500a47aa8fae2a664d", "generation_seconds_batch": 24.844639451010153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6565967798233032, "target_z": -0.0034179530839639374, "ce": 6.1020402908325195, "genuineness": 4.545003414154053, "identity": 0.6283592581748962, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/199455d88ecdeb47f58dfb2a57417781929c4838cc517877e58db0ee7371470f.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "199a642d71cf81644dbea35dc421443594b52cb3ba7a498a09008627f08a7ae6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c59ca3ed729dbb3f5c0a6ed298b48c51d05ace676c99cb300c2af7ee316e2acc", "generation_seconds_batch": 18.024170089047402, "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": 6.022613525390625, "genuineness": 4.281409740447998, "identity": 0.5109595656394958, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/199a642d71cf81644dbea35dc421443594b52cb3ba7a498a09008627f08a7ae6.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "19a8852800d448f9f613741c876dcd63e1788068372a03573f4c7a38197583f3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a7bf49dbe8e6d8ae32e5e52af7cfdd9c81be752a44ce893bdfeb3276fe6e8c4a", "generation_seconds_batch": 23.66000220598653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2979912757873535, "target_z": 1.9687501432323298, "ce": 5.434855937957764, "genuineness": 3.9048471450805664, "identity": 0.554173469543457, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/19a8852800d448f9f613741c876dcd63e1788068372a03573f4c7a38197583f3.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "19d52f77fe87b590f2ae8f823fa00273c0f80a6e7709476443a235a21aa348d8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4b119090940cfb5f29f2b7e0fa4d8d9cb6dd8596855455b013de7acae6826f0d", "generation_seconds_batch": 22.272687204065733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8571786880493164, "target_z": 2.640624928383835, "ce": 5.788304805755615, "genuineness": 3.961331605911255, "identity": 0.5157381296157837, "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_high/19d52f77fe87b590f2ae8f823fa00273c0f80a6e7709476443a235a21aa348d8.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "1a3d15ea9d177294b0fd04d16c96d88a8ee3433d1f9c1c272e72c82e24aa4466", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d292dfdd5cbb549e4edd890ed4cbf86fb23db3bd0770505ebc6e76317751771e", "generation_seconds_batch": 18.40923382504843, "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": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.5086269378662109, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1a3d15ea9d177294b0fd04d16c96d88a8ee3433d1f9c1c272e72c82e24aa4466.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "1aa17f0a84f908b6e7d37a3fb92d6606e41fa7d5fd84bcab76d4a22de13f1556", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "71d288a74534e1a479b801e694ad29e7750568af237c3eaf2d9b360ebffb73a7", "generation_seconds_batch": 21.445947820087895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.633026361465454, "target_z": -0.03173827901199485, "ce": 6.340320587158203, "genuineness": 4.494794845581055, "identity": 0.44484153389930725, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1aa17f0a84f908b6e7d37a3fb92d6606e41fa7d5fd84bcab76d4a22de13f1556.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "1ae308d3670c111af7c1d12b4e8192bfbcf172050c3ec828c245c3ce80265ff1", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d04853e4afae710e53ac13b1fc10915f7208870dbf81ed8132ae650a3294d380", "generation_seconds_batch": 16.76070205704309, "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": 4.381826400756836, "identity": 0.49019116163253784, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1ae308d3670c111af7c1d12b4e8192bfbcf172050c3ec828c245c3ce80265ff1.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "1b2c51d87f2b1c87825c54e4693b4b1a569faacbede3fcca40cc518775a4493d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "535432ba5019cd5f6619ed50fb822146fd95d37b99c7ead3eda621d5ea4ffe68", "generation_seconds_batch": 22.50709457800258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.907860279083252, "target_z": 1.5, "ce": 6.117925643920898, "genuineness": 4.300237655639648, "identity": 0.6787700653076172, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1b2c51d87f2b1c87825c54e4693b4b1a569faacbede3fcca40cc518775a4493d.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "1b30524328adaf5593cc71ffba0abc46784572feb3e8b244e357c76fecf3b1ba", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2d5769818300720cec3478a0e5a27204863c6294fc23053b5b6749c82d33795c", "generation_seconds_batch": 21.835536123020574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.10426139831543, "target_z": 2.9374997135353405, "ce": 5.049636363983154, "genuineness": 3.9676074981689453, "identity": 0.5747535228729248, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1b30524328adaf5593cc71ffba0abc46784572feb3e8b244e357c76fecf3b1ba.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "1c1817f474d4e92babe8de97bee30ba78d8515d72f2991f584599ae9387bd654", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ad63b3781cea2c04f6c9ee009ba63704b84ee7a7754265b97fc1d4ee0df44394", "generation_seconds_batch": 16.95107209100388, "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": 6.213237762451172, "genuineness": 4.149612903594971, "identity": 0.5303215980529785, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1c1817f474d4e92babe8de97bee30ba78d8515d72f2991f584599ae9387bd654.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": "1c34f0093df7de838ed199076dc6662c8b427f93aad70705a4f215f20b9a2e65", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1124daab305f0a78aea6d2e287201960646b2c25a0307319f25efafc3a35538", "generation_seconds_batch": 22.51838194206357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.206960678100586, "target_z": 1.8593750716161648, "ce": 6.403861999511719, "genuineness": 4.268857479095459, "identity": 0.5912784337997437, "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_high/1c34f0093df7de838ed199076dc6662c8b427f93aad70705a4f215f20b9a2e65.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "1c364e3e7400a5b15fd05edd248f26010d4807931300dfff437e887de3164403", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "dfc11cc1b147eb6c18b45ed61a03273de7e69fd6e99991294c155d75dae5498c", "generation_seconds_batch": 23.229214551043697, "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": 6.38003396987915, "genuineness": 4.394378662109375, "identity": 0.5398409366607666, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1c364e3e7400a5b15fd05edd248f26010d4807931300dfff437e887de3164403.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "1c91af0c466fd29d09fcdd1cdb98c05e2fa497bb62e282867ee0315c5a8d94ac", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ac311a26e8545e4fcc01a69bdd1e4c5a49404a2a5929660c37402ace98c1866d", "generation_seconds_batch": 15.907733820029534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8089916706085205, "target_z": 0.17968753580808242, "ce": 6.467403411865234, "genuineness": 4.7458367347717285, "identity": 0.34330296516418457, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1c91af0c466fd29d09fcdd1cdb98c05e2fa497bb62e282867ee0315c5a8d94ac.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": "1ca947ba6f0fb496a1f475e1049be97571a92217f6a0f8dbcf8013aa54170cab", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e28b7b1f062cef3f1d54e221e73c4eef16365e8eb3c205cfc7bb3ab597418a76", "generation_seconds_batch": 17.793280541081913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.303157329559326, "target_z": 0.7734373925757527, "ce": 5.998785495758057, "genuineness": 4.6705241203308105, "identity": 0.6182804107666016, "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_high/1ca947ba6f0fb496a1f475e1049be97571a92217f6a0f8dbcf8013aa54170cab.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "1ce60724164c4bcf8103224d936c5bff3e61ef49d51541f1a5e8edb8e1f9462c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "946e47b6bc2ad4439c9b796e70303e5f5b17f652409306ea1fee7245d9fcd6ad", "generation_seconds_batch": 23.450546634034254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.067453384399414, "target_z": 0.490234419760103, "ce": 6.23706579208374, "genuineness": 4.833701133728027, "identity": 0.699164628982544, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1ce60724164c4bcf8103224d936c5bff3e61ef49d51541f1a5e8edb8e1f9462c.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": "1d34725f292fdeb12c501f0f496fb5ddda4cfcfc0bdd7eb0368a657b85066361", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3f58bcd0eab9be18d0743f80be4c2792b62527d94a12ba0e52f2d992bb93498", "generation_seconds_batch": 17.793280541081913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4429543018341064, "target_z": 0.9414062320959588, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.6235702037811279, "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_high/1d34725f292fdeb12c501f0f496fb5ddda4cfcfc0bdd7eb0368a657b85066361.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "3773938b525eddbccb29a1ce0d6a7fad60e97421e5a2af10f7857b09fb1f64b6", "generation_seconds_batch": 17.45894847298041, "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": 6.229123115539551, "genuineness": 4.319066047668457, "identity": 0.5453389286994934, "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_high/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "248d8d85c2e9c87ea22dceaaef50e96d3255c77d0c1d2a687aa16f638e3bf159", "generation_seconds_batch": 15.729415972949937, "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": 6.1894097328186035, "genuineness": 4.375550270080566, "identity": 0.5710752010345459, "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_high/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "1db88a695d2aa59190401ebb0c3023228763f7bb12108bc3673f61e9057a5fd1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2a461f6324ed17eb9ed96fea8d4d4074ad369dbb6570fd52109a19517e8c8b44", "generation_seconds_batch": 22.859268696047366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 6.0305562019348145, "genuineness": 4.168440818786621, "identity": 0.49318891763687134, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1db88a695d2aa59190401ebb0c3023228763f7bb12108bc3673f61e9057a5fd1.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "1df01af76602f87b531264b27b3e63d566b5c79398b152b140229d4280f4cfce", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b6f0c90bce42a5f576ba083ce0e8ed8494c17857ce356a261a65d3d05b75c9ed", "generation_seconds_batch": 24.132291703950614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8883538246154785, "target_z": 1.4765626074242473, "ce": 6.260893821716309, "genuineness": 3.9550557136535645, "identity": 0.43893375992774963, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1df01af76602f87b531264b27b3e63d566b5c79398b152b140229d4280f4cfce.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "1e4fce172c015ddbb35c898d03e0423193528b874d5877d9aa6a8751d4a65f9d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "80bc40aff3898961818923314b6c4b45559ca1b2bc9b5a4f54664e227a6c0490", "generation_seconds_batch": 22.567111302050762, "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": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.578616738319397, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1e4fce172c015ddbb35c898d03e0423193528b874d5877d9aa6a8751d4a65f9d.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "a61ca3a29a1266b21ba52591b01dda6324497e0e3bc9b82930e223400352539f", "generation_seconds_batch": 17.45894847298041, "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": 6.125868320465088, "genuineness": 4.036643981933594, "identity": 0.6230457425117493, "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_high/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "1f60a5e8ae9b5b3b91397a33ca4dfe31a497976e8a97496414015636f7244087", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f03e96eef0efe632cd53a3282c288b168588d5c960bfc31107e4ab2aedb9c222", "generation_seconds_batch": 21.46896954299882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9988908767700195, "target_z": 1.6093750716161648, "ce": 6.197352409362793, "genuineness": 4.356721878051758, "identity": 0.6876267194747925, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het Oude Raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1f60a5e8ae9b5b3b91397a33ca4dfe31a497976e8a97496414015636f7244087.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "1f8b5edb827d8790ce4125c6f3acaf853a5601e3b0aa0bba9f6986da49655103", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "83a7cb6c901d7cc398ca715ffb6d2a296ee73cb09459aae7b5c9ecca1ba4b736", "generation_seconds_batch": 23.12853293691296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7791526317596436, "target_z": 2.546875071616165, "ce": 5.315715789794922, "genuineness": 4.030367851257324, "identity": 0.6084141731262207, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1f8b5edb827d8790ce4125c6f3acaf853a5601e3b0aa0bba9f6986da49655103.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "1fa380ac61dea42c02a3f873ce5eb64a1ed84191a60f562615bb54d47bc0dab6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f0136a03454aa1a67b19c5ab9feec0d663e4ba301a6e256140c28206e50b50c", "generation_seconds_batch": 25.47056671709288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5970914363861084, "target_z": 2.328124928383835, "ce": 5.831989288330078, "genuineness": 4.005263805389404, "identity": 0.4413754343986511, "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_high/1fa380ac61dea42c02a3f873ce5eb64a1ed84191a60f562615bb54d47bc0dab6.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "1fe84ad453a204fe737ad4f6fa959744d9b0150925dc3335614164a13eb2fd87", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "38ff3c6932081f19cba8ed08cce60247e5a85aca7fcd42f22140371bd179e0c2", "generation_seconds_batch": 22.37107196706347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5970914363861084, "target_z": 2.328124928383835, "ce": 6.229123115539551, "genuineness": 4.124508857727051, "identity": 0.43105563521385193, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1fe84ad453a204fe737ad4f6fa959744d9b0150925dc3335614164a13eb2fd87.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "1ffc6ad34cc573a90723be73789402fb4e670d62df178532fe8f5d28961d4f1b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "978d9b7b0de17e5d8f4f82dab3a4824b624f0c975d88b8fcc28c070c3fa8e201", "generation_seconds_batch": 17.86357043602038, "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": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.3625701069831848, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/1ffc6ad34cc573a90723be73789402fb4e670d62df178532fe8f5d28961d4f1b.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": "200866b15cdf60d444f4dd4a50d71f4993035ae3395c8e7de04ec16f95db66e8", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbc39ad0445d68aff995a9b3b70e36336dd67f7cf27bf5dfbeaff616bc861180", "generation_seconds_batch": 17.855276351911016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7843189239501953, "target_z": 1.3515626074242473, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.5628360509872437, "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_high/200866b15cdf60d444f4dd4a50d71f4993035ae3395c8e7de04ec16f95db66e8.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "20acaf50a0efbd9a64679b2ce05576a3954c02ef834e3d10fb2eee4d0d98c520", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f061395005a81321733be6bd3cbf6281c6264f238c7f1fa20c67d46fa2ce9d4", "generation_seconds_batch": 23.85429213603493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 5.919358730316162, "genuineness": 4.469690799713135, "identity": 0.5863605737686157, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/20acaf50a0efbd9a64679b2ce05576a3954c02ef834e3d10fb2eee4d0d98c520.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "20f034efbc6a2ba3961330454c158159d6a793f522c1df202ddd1a33381bf048", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "46cc891f9deb701601b4db1b432c112a39724cfc72cb8d3810872deac776496f", "generation_seconds_batch": 23.596837450983003, "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": 6.006728172302246, "genuineness": 4.432034492492676, "identity": 0.7639172077178955, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/20f034efbc6a2ba3961330454c158159d6a793f522c1df202ddd1a33381bf048.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "215150070edc504a0b521280883c23d7eecc11a99d4b0b2d1bcb2eb5d8be4998", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "65e7394a22d7346c9e8cdf879edc274f5cbe4dd51697ab7a83d4c6f46dff4584", "generation_seconds_batch": 18.799026050022803, "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": 6.205295085906982, "genuineness": 4.6705241203308105, "identity": 0.6064025163650513, "hypothesis": "He encontrado la cajita. Estaba junto. A la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/215150070edc504a0b521280883c23d7eecc11a99d4b0b2d1bcb2eb5d8be4998.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "21a2004233ae4e6296c310b973cbca86721407a6e8bdcd0d51dce66afffcfbe8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "61d0aa375043fdbf71d8f647655b21e18f2475380cad071c7770072a028ce758", "generation_seconds_batch": 23.65791986102704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5980771780014038, "target_z": -0.0737304172758815, "ce": 6.205295085906982, "genuineness": 4.306513786315918, "identity": 0.6155872941017151, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/21a2004233ae4e6296c310b973cbca86721407a6e8bdcd0d51dce66afffcfbe8.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": "21b952474f3957fb5fa0cb96a61473749250ec9bc4e20af9e0989c9e4e60c8be", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccb868a2c38842e8be2d8d35718d6920588ee26e7f1dd02746a59bab896212bf", "generation_seconds_batch": 18.53081536490936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7401394844055176, "target_z": 2.5, "ce": 6.221180438995361, "genuineness": 4.080576419830322, "identity": 0.4983338415622711, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer hebben 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_high/21b952474f3957fb5fa0cb96a61473749250ec9bc4e20af9e0989c9e4e60c8be.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "221bbfac3f3dcf5158ddd856c68445bd9353b78ce2c0bf990f76596c41c9ef99", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ec1433a1435ec264d35098a0038d94856a6b7ddcc6f4127fd956843a6c15a0d0", "generation_seconds_batch": 17.431258630007505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.000226974487305, "target_z": 2.8125002864646595, "ce": 6.356205940246582, "genuineness": 4.011539936065674, "identity": 0.46636369824409485, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude ramen. 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_high/221bbfac3f3dcf5158ddd856c68445bd9353b78ce2c0bf990f76596c41c9ef99.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "2250975ab08dc5f4248578cbac09811f6b2369b467086de19354b92f35d4543d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e5eb5af207a8fa11e1971fd07fcb8b8067ab4bbb5ea5c5c105a395bdca37ceb7", "generation_seconds_batch": 23.666904239100404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.278484582901001, "target_z": 1.9453124641919175, "ce": 6.078212261199951, "genuineness": 4.243753433227539, "identity": 0.48427748680114746, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2250975ab08dc5f4248578cbac09811f6b2369b467086de19354b92f35d4543d.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "226421b2bfa8a66c23b236423d67997587774f95e7d6a33322edcfc0e65f15a3", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7604393cc86ebfa2691a6a6e685fc8bbaa579ec64af106f0833d14edacf9a7d6", "generation_seconds_batch": 16.828247010009363, "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": 6.181467056274414, "genuineness": 4.3881025314331055, "identity": 0.5705279111862183, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/226421b2bfa8a66c23b236423d67997587774f95e7d6a33322edcfc0e65f15a3.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "23b4021ba77aead8dde2d47416da61673ae85ea2f03c717ea378b969f48f37c5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0159e48df38c2ba7d4093b8263cd0b53963a4dd6645b3bdd43c9bd6ba281b11e", "generation_seconds_batch": 22.72342200600542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5580782890319824, "target_z": 2.2812498567676704, "ce": 6.2767791748046875, "genuineness": 3.886019229888916, "identity": 0.436410129070282, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/23b4021ba77aead8dde2d47416da61673ae85ea2f03c717ea378b969f48f37c5.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "243bb0889e29745ff1927e3f2098be2599c22e47f91aaf1f79b991927dda8f85", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6ec0155ea99fd5dcfcf5469880c33120ccdb4727ac9d6493ebba144f59538a74", "generation_seconds_batch": 23.713894524029456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.610095977783203, "target_z": 2.3437501432323296, "ce": 6.356205940246582, "genuineness": 3.9676074981689453, "identity": 0.5237455368041992, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/243bb0889e29745ff1927e3f2098be2599c22e47f91aaf1f79b991927dda8f85.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "249f8bfe5f1a230d1153b46632b22dc9ca3e1ebac10c21f8b909b774daecf777", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5be3360acf908fca660e9e87532c057db02a48b5d52892b4abfa347fa5a1bb29", "generation_seconds_batch": 23.666904239100404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7648122310638428, "target_z": 1.3281249283838352, "ce": 6.141753673553467, "genuineness": 4.143336772918701, "identity": 0.6444190740585327, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/249f8bfe5f1a230d1153b46632b22dc9ca3e1ebac10c21f8b909b774daecf777.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "24e978ff54f83ac9f099d45e72dc6a59f05838558d435b895c3aa9789a580ed3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ab9d43741b0bdbd334eda1e98689d1a070e7bd9dc080ba898f57ebe408001148", "generation_seconds_batch": 25.742135660955682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.650444984436035, "target_z": 3.5937501432323296, "ce": 5.95112943649292, "genuineness": 3.71028995513916, "identity": 0.32742035388946533, "hypothesis": "De deur staat nog open. De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.2631578947368421, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/24e978ff54f83ac9f099d45e72dc6a59f05838558d435b895c3aa9789a580ed3.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "2524290823a6a830d6f2e8f2ec37651a609a1cf7798323c9e8498947c91ae4ae", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "18e76e71091fb21605ccd9d2bb9509da0eebc9c8b1a1734f815d39e0ca63b812", "generation_seconds_batch": 25.796130744973198, "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": 6.507116317749023, "genuineness": 4.65797233581543, "identity": 0.6990156173706055, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2524290823a6a830d6f2e8f2ec37651a609a1cf7798323c9e8498947c91ae4ae.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "25d8c23de082731ca0894bfafe2e560b8451bc29db1446b85a830777dc26b05d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fb96d241c30648cf23a927a3cd01dc1f32c71144d89dd3d3bbeae6477d4ceada", "generation_seconds_batch": 22.538426212966442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 6.117925643920898, "genuineness": 4.111956596374512, "identity": 0.6173728704452515, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/25d8c23de082731ca0894bfafe2e560b8451bc29db1446b85a830777dc26b05d.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "25e201fc889fc430322326cc087d3db64eab9ed401d7beccedfbaffc1c35e72f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "89a54ad4034126e648cab89e90b236473dabd9ec2a48e9ab4f440c012a551cb7", "generation_seconds_batch": 16.47344362095464, "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": 6.1020402908325195, "genuineness": 4.419482707977295, "identity": 0.6327237486839294, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/25e201fc889fc430322326cc087d3db64eab9ed401d7beccedfbaffc1c35e72f.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "25edc35a855f6d8a79ea01517d5806641675a45d9961a011724c8e875f3031d7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "859dc9ed6c07c8f1622b383a73ebb9312464a860b3dbab0738624e7929d03224", "generation_seconds_batch": 17.431258630007505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0379037857055664, "target_z": 1.6562498567676702, "ce": 6.268836498260498, "genuineness": 4.281409740447998, "identity": 0.6396652460098267, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/25edc35a855f6d8a79ea01517d5806641675a45d9961a011724c8e875f3031d7.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "25fb52ec56998147730908147a82c0d1dbd749ae2e26181e0ab546bbca9b60e6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ba197aa42c68cf0bd11ef8ffc583d2ac3d203a6282f4c104c462a505b66e6267", "generation_seconds_batch": 24.426989419036545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6265242099761963, "target_z": -0.03955074320391242, "ce": 6.149696350097656, "genuineness": 4.770940780639648, "identity": 0.567916989326477, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/25fb52ec56998147730908147a82c0d1dbd749ae2e26181e0ab546bbca9b60e6.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "261968fe64fb5c9759d62d4c90ec397d6a6e23ac0069e91c21ac366c4a8da814", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8def421775afab21f03b71dfc515c495dd8b4889edb2ee6d726754be9d7e1b45", "generation_seconds_batch": 23.959679967956617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4884696006774902, "target_z": 0.9960937679040412, "ce": 5.9431867599487305, "genuineness": 4.519899368286133, "identity": 0.6783190965652466, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/261968fe64fb5c9759d62d4c90ec397d6a6e23ac0069e91c21ac366c4a8da814.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "2657da6fb997c8f69fba52f3fdd16844d29fff7227f7730aa6dc5202778e640b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3b738df08aedd9097340c4d8d166a10a70585616117597578e6cebaf08f8eb67", "generation_seconds_batch": 23.44189335696865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2654802799224854, "target_z": 1.9296875358080825, "ce": 6.0146708488464355, "genuineness": 4.093128204345703, "identity": 0.5977988839149475, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2657da6fb997c8f69fba52f3fdd16844d29fff7227f7730aa6dc5202778e640b.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "2667a43aa4d00f8ae0a8eaf06e27f295a7ccf7b16c05f22e45a2e2c43cb27d0a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "38f9ec768fa6a9b0d91398e6bcf00b399cfa12d6498a5609edc813ee618fe44b", "generation_seconds_batch": 16.67468790698331, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.013810396194458, "target_z": 0.42578130371212364, "ce": 6.23706579208374, "genuineness": 4.193545341491699, "identity": 0.4914698302745819, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2667a43aa4d00f8ae0a8eaf06e27f295a7ccf7b16c05f22e45a2e2c43cb27d0a.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "269cb510021ddd0d806ecd6fe3630bb7e5811f3bc7a7001983cfe46b41446e24", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5b0caf589aaa1272d5eb830ac3376a3638cf64a291f15a43c9644ba9af0a46d3", "generation_seconds_batch": 24.93322793301195, "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": 5.95112943649292, "genuineness": 4.595211505889893, "identity": 0.7405129075050354, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/269cb510021ddd0d806ecd6fe3630bb7e5811f3bc7a7001983cfe46b41446e24.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1d4507804f00ba9814ce77fae19709b279f3778bb2c8dfd1eeb1c7a036533121", "generation_seconds_batch": 18.90785944298841, "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": 6.300607204437256, "genuineness": 4.20609712600708, "identity": 0.4912691116333008, "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_high/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "26f5e2eda9d7ae028369954b98d92c93735f4e940d2953deabbcdb949103eee4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1b50914a8e0446ce9714fa5d8d547d0290255dd4e0d43939b2fd3584ec9db084", "generation_seconds_batch": 23.88750126992818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.491720676422119, "target_z": 1.0, "ce": 6.205295085906982, "genuineness": 4.287685871124268, "identity": 0.5883534550666809, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/26f5e2eda9d7ae028369954b98d92c93735f4e940d2953deabbcdb949103eee4.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "27839f818a04659e872a5c6776ff0f0e581a00c01e598797f8920fdcf159e5e9", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7e7b09ef002cdd700688831dce946249002fe3c116eceb08428021fdc6a14205", "generation_seconds_batch": 15.188824643963017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7618508338928223, "target_z": 0.12304688395202061, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.7693273425102234, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/27839f818a04659e872a5c6776ff0f0e581a00c01e598797f8920fdcf159e5e9.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "27a5223da6ac92c81e5916cee9b4f8255a062318b11b21ad2237bcdaced77b05", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "283ca50a1c4268f651544c12750bdd832b2befad33dffd711ed96b2771edbb42", "generation_seconds_batch": 17.214929033070803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.641270875930786, "target_z": 1.1796875358080825, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.533179759979248, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/27a5223da6ac92c81e5916cee9b4f8255a062318b11b21ad2237bcdaced77b05.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "27a7c237f1e9fe684ffa943846ca1e7f7d3059f41dd6d2ca0460270cdbfc19a3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "749bd48a3f23a7746b6a188c3e105f675acaec9b2fdaedca907bbd59189db5c8", "generation_seconds_batch": 21.237716658972204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2264671325683594, "target_z": 1.8828124641919175, "ce": 5.506340026855469, "genuineness": 4.105680465698242, "identity": 0.5840061902999878, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/27a7c237f1e9fe684ffa943846ca1e7f7d3059f41dd6d2ca0460270cdbfc19a3.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "28060c76bf1be76d65beb6076ec1d1f0f1731b611fa80f6e86e5a05492f9e8dd", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "72954ff4e7f572627b58b7002d762b9daa61aeee37109c3e343898bc5f15d7f3", "generation_seconds_batch": 17.055181364994496, "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": 5.824047088623047, "genuineness": 4.20609712600708, "identity": 0.5548300743103027, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/28060c76bf1be76d65beb6076ec1d1f0f1731b611fa80f6e86e5a05492f9e8dd.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "2834b4c2f4bb9c6ad11bb2c236d3ff76a2da8a16f452f3f9fde264cefd43ea70", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "6ed69daa0f3c5d24dd2c88871b9ebda1b90f36ded6e80a02c585ff7ac2ab3888", "generation_seconds_batch": 22.934322233079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5190653800964355, "target_z": 2.234375071616165, "ce": 5.681078910827637, "genuineness": 3.9550557136535645, "identity": 0.5605082511901855, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2834b4c2f4bb9c6ad11bb2c236d3ff76a2da8a16f452f3f9fde264cefd43ea70.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "2880c6eab5dc943a6591ff5f002130c5c9427763efa4341389644eb83bc60a2d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2bbf3d06c68deb262ad626cbd2fbb3de77635ed08f653aef9921e1948d7e5621", "generation_seconds_batch": 23.003163064015098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3630130290985107, "target_z": 2.046875071616165, "ce": 6.141753673553467, "genuineness": 4.20609712600708, "identity": 0.3016311526298523, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2880c6eab5dc943a6591ff5f002130c5c9427763efa4341389644eb83bc60a2d.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "28fba32e6792c1030f13040693e87e84e361c9a60f5e72180648ed5ed64db9ad", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "55b99f36f0bb8899deea61b5654d7167844ade8189dd0e3321c57b45d18a8927", "generation_seconds_batch": 24.210008699912578, "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": 6.1020402908325195, "genuineness": 4.369274139404297, "identity": 0.7138455510139465, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/28fba32e6792c1030f13040693e87e84e361c9a60f5e72180648ed5ed64db9ad.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "296079337c80e39667a6811b0c6c1068177a2ac5542a73f87a0280194aa1924d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "95a05b1507a085379243aef87ada99cd820a3900f5f178c35859400e25b41fb3", "generation_seconds_batch": 16.22692346200347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7453057765960693, "target_z": 1.3046875358080825, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.6700016856193542, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/296079337c80e39667a6811b0c6c1068177a2ac5542a73f87a0280194aa1924d.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": "2970ea57dce0d4567f22c33fe96b2f84d536944baabcb80438d4b6da2c5a6e66", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6a6471011369047302fa6eeea0111d870065dbe1e9be412253ecd91ea9e0dfa", "generation_seconds_batch": 16.316368971951306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892534255981445, "target_z": 1.1171875358080825, "ce": 5.784333229064941, "genuineness": 4.287685871124268, "identity": 0.7407276630401611, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Pon-les quand tu partiras.", "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_high/2970ea57dce0d4567f22c33fe96b2f84d536944baabcb80438d4b6da2c5a6e66.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "298039974454ca056449adfb751ca51cdc0d7b9fa4a93f0f247044bf96843d65", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "73dfa9c7c9fdaa8c8c7197096101653f3d4b8fee877213f2db5a0ac544084c62", "generation_seconds_batch": 22.809880690998398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.93740975856781, "target_z": 0.333984419760103, "ce": 6.284721851348877, "genuineness": 4.595211505889893, "identity": 0.7921470403671265, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/298039974454ca056449adfb751ca51cdc0d7b9fa4a93f0f247044bf96843d65.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": "29918807cc5ac86be7381c1702aadae8dae523bc5b21908a373d61b149073a44", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c89abfc2ec73f850cc7153e4b92acf5a9928658d93bf0ec438860c91085d6ace", "generation_seconds_batch": 22.96548067999538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3630130290985107, "target_z": 2.046875071616165, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.6797508597373962, "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_high/29918807cc5ac86be7381c1702aadae8dae523bc5b21908a373d61b149073a44.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "299d1d085cf2fa225537e45ee03695e5d52c1b45e5378eada596ae191ccc8778", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "94a1d023fed0fd726924be04c96de441d41f8d89e09978278b39010e5a7ac4d5", "generation_seconds_batch": 18.1167323050322, "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": 6.260893821716309, "genuineness": 4.557555198669434, "identity": 0.5942307710647583, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/299d1d085cf2fa225537e45ee03695e5d52c1b45e5378eada596ae191ccc8778.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "29f166e2647060deb89ccdda730e3e641cec5ab0bfcd0ad377b6c2b9cc59ca01", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ca2740d3944a40747d9c2bd6b805982804c69ae92aaf4715924795dfdb49025c", "generation_seconds_batch": 25.435089610051364, "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": 6.022613525390625, "genuineness": 4.519899368286133, "identity": 0.6368138790130615, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/29f166e2647060deb89ccdda730e3e641cec5ab0bfcd0ad377b6c2b9cc59ca01.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": "29f4e5d9fd69d01eb7b41d7682fca34a499090961396fc37a989bd7ec376d422", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b32a023e1371d64ba8309e9c87493abb3b76e85b2cf738f0c672ca3227e0697", "generation_seconds_batch": 22.117929962929338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.200458526611328, "target_z": 1.8515626074242473, "ce": 6.197352409362793, "genuineness": 4.243753433227539, "identity": 0.5430923700332642, "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_high/29f4e5d9fd69d01eb7b41d7682fca34a499090961396fc37a989bd7ec376d422.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "2a3b84d6d1083c0b268edc1ec502dc235ae1742a156c5293e210e79a64dbdc40", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "597fb53711806ba6f55295fe06c2294b771c3984bda2e980d6ff44644e0f1dd8", "generation_seconds_batch": 18.021841699024662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.124347448348999, "target_z": 0.5585937679040412, "ce": 6.3641486167907715, "genuineness": 4.093128204345703, "identity": 0.37204837799072266, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2a3b84d6d1083c0b268edc1ec502dc235ae1742a156c5293e210e79a64dbdc40.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": "2af3325b34c5fecadf6241a87b957c27a930aa70561c9d38b634df82dc797d60", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3647ccb11f2c3b25be10daaa162c6b7fac6f9dc0efdf8a0d73d800e0364471c5", "generation_seconds_batch": 22.51838194206357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.688122034072876, "target_z": 2.4375, "ce": 6.356205940246582, "genuineness": 4.193545341491699, "identity": 0.7974886894226074, "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_high/2af3325b34c5fecadf6241a87b957c27a930aa70561c9d38b634df82dc797d60.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "2b026b58cb075646cd050f4cde004192087d9c67be9dcd0fe6dafb3d3589e740", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7c7ee80a63687415d78cffd8b8dd68b30cce41969bd97c96546f5f407908cda4", "generation_seconds_batch": 23.493406652007252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.117845296859741, "target_z": 0.5507813037121236, "ce": 6.1894097328186035, "genuineness": 4.193545341491699, "identity": 0.21208199858665466, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2b026b58cb075646cd050f4cde004192087d9c67be9dcd0fe6dafb3d3589e740.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "2b73bf3b90b91851be235ac5d2d125c23728dc088ffacc5382e0c2c9c8da4e2e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1ca3caec85decd181c0fa46bfdd34500d05b7cbda8801282663240353b55b081", "generation_seconds_batch": 15.622364434995688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.907860279083252, "target_z": 1.5, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.3401312232017517, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2b73bf3b90b91851be235ac5d2d125c23728dc088ffacc5382e0c2c9c8da4e2e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "2c0abe7a2a0d8649a88048eea001d81860c618db8823640c2885ccf89caa8a5e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "a307989ee107db40871ec1a549df2c6a999087439e25580ccad49761bcd5c997", "generation_seconds_batch": 22.68887325504329, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0379037857055664, "target_z": 1.6562498567676702, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": 0.32687103748321533, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2c0abe7a2a0d8649a88048eea001d81860c618db8823640c2885ccf89caa8a5e.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": "2c35100cc9afd062226f4726248f2cc87298d9c3a29c6368be9b6a1f6c809ba0", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "574810f346b160f4369f9640f62a1a7350c91862d64522882554dcb59f3684fe", "generation_seconds_batch": 17.465600084979087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.232969284057617, "target_z": 1.8906249283838352, "ce": 6.205295085906982, "genuineness": 3.936227321624756, "identity": 0.4865512251853943, "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_high/2c35100cc9afd062226f4726248f2cc87298d9c3a29c6368be9b6a1f6c809ba0.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "2d36ebbd5358b1e18e788e6f8bd500c55d60bb02a50392db243b44b0b488cd13", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6bcae2e6d0736dbc64d6fe69f1bda471a8a91f76b5339616a39ab4792954f822", "generation_seconds_batch": 16.948556926916353, "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": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.321488618850708, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2d36ebbd5358b1e18e788e6f8bd500c55d60bb02a50392db243b44b0b488cd13.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "2da1420573f33ad01f634174d2aa662a251f94d2307d181b31d7b85c82a80281", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2b5e5be3141b5bae4a3a913d3b8af69c24323c0c7a86efb061e708551dd0c0e7", "generation_seconds_batch": 17.65382483799476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.413694381713867, "target_z": 0.9062498567676703, "ce": 6.356205940246582, "genuineness": 4.369274139404297, "identity": 0.6789231300354004, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2da1420573f33ad01f634174d2aa662a251f94d2307d181b31d7b85c82a80281.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "2dcb1cc8825cac71d752882983a8b8f4458dfeabdd314193737ea391c0ff344e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "46a6a509504efc3408bb19fead3ca676e3893f63196398c308669c2a9f652144", "generation_seconds_batch": 22.050309470971115, "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": 6.2767791748046875, "genuineness": 4.909013748168945, "identity": 0.6503425240516663, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2dcb1cc8825cac71d752882983a8b8f4458dfeabdd314193737ea391c0ff344e.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "2e7acfe654dd5e25a711fb71fd6f5473a6ed5296a0f114f31b0d433b078d87d3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f2d9c782783c648665f56891a1a36ccc1680620a56882d4123b9994f3514a298", "generation_seconds_batch": 23.11722616502084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1145942211151123, "target_z": 0.5468750716161649, "ce": 5.538110733032227, "genuineness": 4.344170093536377, "identity": 0.5840466022491455, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2e7acfe654dd5e25a711fb71fd6f5473a6ed5296a0f114f31b0d433b078d87d3.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "2ea7c111ce2c15ae8a8b1ca2c6b30608f8732340fde3ad960a4b6536c3981380", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f2c93ada808f819ed77082be2f8f0b758ae40bd1f9f80ae8e92df544ffb41191", "generation_seconds_batch": 17.259381529060192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5220829248428345, "target_z": -0.16503902221590727, "ce": 6.332377910614014, "genuineness": 4.306513786315918, "identity": 0.592809796333313, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2ea7c111ce2c15ae8a8b1ca2c6b30608f8732340fde3ad960a4b6536c3981380.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "2ee7a6c5677e516690faf18e0882da224af1046d983e43ade4691a0130f0fef5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "554499e3d35c36060ff070fb5638740407cedc0fa4c6b8a473cdbb35cac6d64b", "generation_seconds_batch": 22.525076239951886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5580782890319824, "target_z": 2.2812498567676704, "ce": 5.796247482299805, "genuineness": 4.356721878051758, "identity": 0.5534032583236694, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2ee7a6c5677e516690faf18e0882da224af1046d983e43ade4691a0130f0fef5.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": "2f64f1bfa5e4ec9f6566a4fe2fe2b393799723560542344450d666cab8f29cfd", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86866be48378a305f0b881497019f99255c7900fa3240eaf493ed92e16118bfc", "generation_seconds_batch": 23.555441918899305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5580782890319824, "target_z": 2.2812498567676704, "ce": 6.284721851348877, "genuineness": 4.049196243286133, "identity": 0.4885820150375366, "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_high/2f64f1bfa5e4ec9f6566a4fe2fe2b393799723560542344450d666cab8f29cfd.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "2fd75d4bd72f6b8ac5bb7341662abb4d3ca76e3e78bba161c718d6c58cbfd0c6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "be06b62276926ad147c91952c2127a7787fc93bfc31f302af64b30f641deb3ea", "generation_seconds_batch": 22.707033994956873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6958130598068237, "target_z": 0.04370118642203349, "ce": 6.3641486167907715, "genuineness": 4.444586753845215, "identity": 0.44104552268981934, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/2fd75d4bd72f6b8ac5bb7341662abb4d3ca76e3e78bba161c718d6c58cbfd0c6.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "30135639b7b6152a1c00c67954784d76531f4dfcecd56bb57a8b8ba8a88481c5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2ee94a6ccd8683b816cf0911073f350cb5d743053a248a5a2454b8295dd07975", "generation_seconds_batch": 24.20763007795904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.766148090362549, "target_z": 2.5312498567676704, "ce": 6.09409761428833, "genuineness": 4.174716949462891, "identity": 0.3663918972015381, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/30135639b7b6152a1c00c67954784d76531f4dfcecd56bb57a8b8ba8a88481c5.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "30949a9e31578a43f7399f074cb07eb59cde2503a07bda5e5e1405ead5bdcdaa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cd3a9e44b37d08229806c037e3bf995c5444022efbdf0f826c7cb0bdfd4ea0cc", "generation_seconds_batch": 16.436631727963686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1484408378601074, "target_z": 1.789062320959588, "ce": 6.038498878479004, "genuineness": 4.149612903594971, "identity": 0.3924064338207245, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/30949a9e31578a43f7399f074cb07eb59cde2503a07bda5e5e1405ead5bdcdaa.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "3157008878ae9f604e989d3414f2f2d554cf89abd0ae181abd5d6081d0209567", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fc3f115166b41c2dac5a6a81080e508a864cd42bb53e0672bd3c0f4b1ac3d78e", "generation_seconds_batch": 21.999378002947196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9742178916931152, "target_z": 2.7812498567676704, "ce": 5.748591423034668, "genuineness": 4.15588903427124, "identity": 0.5142133235931396, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3157008878ae9f604e989d3414f2f2d554cf89abd0ae181abd5d6081d0209567.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "31f4549f31a96f62a9d6296fbd96d790401340ea169707ef371179ae7a434501", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c3637edf3945486e983c7bd1943da2af74b68f6b4634cc3423c8c24abfc7ee2e", "generation_seconds_batch": 22.37035262794234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2199649810791016, "target_z": 1.875, "ce": 5.788304805755615, "genuineness": 4.243753433227539, "identity": 0.15180645883083344, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/31f4549f31a96f62a9d6296fbd96d790401340ea169707ef371179ae7a434501.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "3217921612a781323e79f2cc58c64aa15146face7a128ae102ef2ed059cbb574", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "57ecbc301f310f43b0c0b3ebedcee24e3a01e5eccca481bb6a713f96f7ac2daf", "generation_seconds_batch": 17.909289096016437, "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": 6.38003396987915, "genuineness": 4.6705241203308105, "identity": 0.5579779148101807, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3217921612a781323e79f2cc58c64aa15146face7a128ae102ef2ed059cbb574.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3291e734b0578bb220124c92dcdb678e1bb05c4bbcc87dfd6e3ad3fed33dbd82", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5da28e18c83f0ce055ae61d902387a60b14664d36a2cbe3f912fb0f8147da82f", "generation_seconds_batch": 16.66556944604963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7151163816452026, "target_z": 0.06689451110795364, "ce": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.7049800157546997, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3291e734b0578bb220124c92dcdb678e1bb05c4bbcc87dfd6e3ad3fed33dbd82.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "32e6f04194063997d3434aeb8469e98b93ee49ce7f2dbb4ec5c2d3fe79fe7bcb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "cc5edda0e25b8bc73a0ba716839c84fd03f4667c06be7b40c1533347f2950d5f", "generation_seconds_batch": 17.208584030973725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2836508750915527, "target_z": 0.75, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.624194324016571, "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_high/32e6f04194063997d3434aeb8469e98b93ee49ce7f2dbb4ec5c2d3fe79fe7bcb.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "33a5515da146a2d129fb96ca544f5b729c523f12b78b6b464f9694bf0e13e8c4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f08de6f00140a711aeac5b0fafeabb0e1335184ee991071c85aeaedd0775c3e9", "generation_seconds_batch": 23.003163064015098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5632448196411133, "target_z": 1.085937679040412, "ce": 6.133810997009277, "genuineness": 4.607763767242432, "identity": 0.7234219312667847, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/33a5515da146a2d129fb96ca544f5b729c523f12b78b6b464f9694bf0e13e8c4.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "51431f69eb74b02c3a85b3e96c09eb0144797f135a1720665f13e2160194a17f", "generation_seconds_batch": 17.23343314800877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 6.046441555023193, "genuineness": 4.20609712600708, "identity": 0.41428709030151367, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder. Tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "2041e066b0dee15c2b6864ef7827cbb0729f88c7a2c57ff9692a3bb998b3edcd", "generation_seconds_batch": 18.394503772025928, "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": 6.427689552307129, "genuineness": 3.9048471450805664, "identity": 0.2736656963825226, "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_high/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "347f4a618a0e15d10a45b56f4c693adf78baa6c89e343b22eeca258144e57b27", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "104128aed306c5ce37c08c95325fe07860b9d59ba7507950394451b616c0a8a4", "generation_seconds_batch": 25.59293839603197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9988908767700195, "target_z": 1.6093750716161648, "ce": 6.09409761428833, "genuineness": 4.507347106933594, "identity": 0.6806122064590454, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/347f4a618a0e15d10a45b56f4c693adf78baa6c89e343b22eeca258144e57b27.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "3480ace1f9f6672118235c0208de9932f6d1b0bbe79349b5b5f1dc8b475711fa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "af819b2ce6978833e270d7621e4bf50e290ccf8d98b7c5eede5d217287d8b3bc", "generation_seconds_batch": 29.32788363506552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.909196138381958, "target_z": 2.703124928383835, "ce": 5.303802013397217, "genuineness": 4.118232727050781, "identity": 0.6910003423690796, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3480ace1f9f6672118235c0208de9932f6d1b0bbe79349b5b5f1dc8b475711fa.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "35046d5ef066a79d25858685934ee72a64a2a8235a025c8dfd007c1e33422e94", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c25b3b9b6b59c85a7778f499501e454c0cf99f0dbfe3f775c3684c9d1fc38ff2", "generation_seconds_batch": 22.48021783796139, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.901358127593994, "target_z": 1.4921875358080825, "ce": 5.681078910827637, "genuineness": 4.293961524963379, "identity": 0.7034339308738708, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/35046d5ef066a79d25858685934ee72a64a2a8235a025c8dfd007c1e33422e94.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "355b28a29fc3bf8652342ae239a948db1a2eaadf020b081b645c211b19fc684c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "07fc9ab7f522715dc0882733345c64b4f69b24000399f87befd10a3f11c6ad0d", "generation_seconds_batch": 21.599320476991124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.052244186401367, "target_z": 2.875, "ce": 5.4388275146484375, "genuineness": 3.9550557136535645, "identity": 0.5484358072280884, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/355b28a29fc3bf8652342ae239a948db1a2eaadf020b081b645c211b19fc684c.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "35ad1cac7ec4f2816e6e6610439b7f8988bfc1c29dbefc9a11460a47c42a117e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2f587d74cc8d6cd8c7c39dc974998a3b3191388a78c6919724997b9cf3a3664e", "generation_seconds_batch": 23.65791986102704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9764227867126465, "target_z": 0.3808593481439382, "ce": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.38380134105682373, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/35ad1cac7ec4f2816e6e6610439b7f8988bfc1c29dbefc9a11460a47c42a117e.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "35ddacfa125bc5bbd44c96eab477ed124797a0172a73e4e97d0b362d07ee025b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "47204310c83b18fe57e312baa5dfca128d65c9487b7c42c02b133a627a7227bd", "generation_seconds_batch": 22.16238859493751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.429370880126953, "target_z": 3.3281252148484946, "ce": 5.33954381942749, "genuineness": 3.4561104774475098, "identity": 0.3066712021827698, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/35ddacfa125bc5bbd44c96eab477ed124797a0172a73e4e97d0b362d07ee025b.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "3633de5da10097488ebcbb1a298267d1ed55d3345082cae40c2e2d2f7c7b1746", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f9c16736b05fae1bacb72f2b6c9b7676e51e834c654396c9d4bf6918ed0d7409", "generation_seconds_batch": 26.047198134008795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1484408378601074, "target_z": 1.789062320959588, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.6287803053855896, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3633de5da10097488ebcbb1a298267d1ed55d3345082cae40c2e2d2f7c7b1746.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "367b8180c18e3c8ecd4d67c210642deaa1676c74bca28c2366e34279353b97d0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0e0901acd697c9be06f2649e0c2950a30ba03dd33d8a582808867e548e95400c", "generation_seconds_batch": 18.287825005943887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868847131729126, "target_z": 1.4531249283838352, "ce": 5.839932441711426, "genuineness": 4.469690799713135, "identity": 0.5537667870521545, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/367b8180c18e3c8ecd4d67c210642deaa1676c74bca28c2366e34279353b97d0.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "36d737bfb4d09f1d701f1458b5947e045b3cecc4af0280635e1068cafa859917", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5a15f9238935ca507eb157bf16c412472cb1236145ded3eb15bfcc2430f708ce", "generation_seconds_batch": 24.200259198085405, "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": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.5254401564598083, "hypothesis": "Die Tür 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_high/36d737bfb4d09f1d701f1458b5947e045b3cecc4af0280635e1068cafa859917.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3704895450d8956b77435f1799c5f61b2a4a998f400639996f8e28802deee3be", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "5b9d894bac6280291889ea9ee78f0240b0ed66c89e8297e1467fd81adde57534", "generation_seconds_batch": 23.280895702075213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8756389617919922, "target_z": 0.259765580239897, "ce": 6.086154937744141, "genuineness": 4.457139015197754, "identity": 0.7426376342773438, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3704895450d8956b77435f1799c5f61b2a4a998f400639996f8e28802deee3be.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "373f4ea8ba63940534f877dc730be25e8f1afd78d5484ba01e7985df3e922790", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0a88d3a257923afa77af727c6e56485cfc60c3b6d58421f553e0d8e39a3d2220", "generation_seconds_batch": 26.047198134008795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940371036529541, "target_z": 1.539062320959588, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.427312970161438, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/373f4ea8ba63940534f877dc730be25e8f1afd78d5484ba01e7985df3e922790.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "3789415872a282eed3c0cb836fee610c1dfc7a84ed5785e2bdf3b722bb36a6eb", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5bbc4a0f917412037df658af0bf866e441112b58b9b0cac2a7102cd6dbf8fd69", "generation_seconds_batch": 15.746108162915334, "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": 6.435632705688477, "genuineness": 4.331617832183838, "identity": 0.40218406915664673, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3789415872a282eed3c0cb836fee610c1dfc7a84ed5785e2bdf3b722bb36a6eb.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "37ab4c851879c45ff69ff746efd003d4c8981781f56cecf3feebf44e198a1367", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b030b982c1de51708baffa1c50e6e579b11d3ea51d97ba2eb0b8ce3887dcafba", "generation_seconds_batch": 23.144915941986255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4754652976989746, "target_z": 0.980468839520206, "ce": 6.125868320465088, "genuineness": 4.595211505889893, "identity": 0.627526044845581, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/37ab4c851879c45ff69ff746efd003d4c8981781f56cecf3feebf44e198a1367.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "388540d67caf4d076f003266e8f882d00504cee2a8a8f45b34dd01213d05d40f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1d530554fcfebd10b79ccaa251b0bd442485b5b7d2779bf372c7e59d3ef6de9b", "generation_seconds_batch": 17.195034984964877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6217641830444336, "target_z": 1.1562498567676702, "ce": 5.6334228515625, "genuineness": 4.432034492492676, "identity": 0.5078308582305908, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/388540d67caf4d076f003266e8f882d00504cee2a8a8f45b34dd01213d05d40f.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": "38b331905a7149579a3825014b3f12a215aa4e4d37169839b241bc19a43b9827", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bc4d9dc2fc374508a23afef64956e5162d4c955eca741541deb615953d15e82", "generation_seconds_batch": 16.690835547051392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9208645820617676, "target_z": 1.5156249283838352, "ce": 6.23706579208374, "genuineness": 4.532451152801514, "identity": 0.4193775951862335, "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_high/38b331905a7149579a3825014b3f12a215aa4e4d37169839b241bc19a43b9827.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "38cb3714acdb405c4e2a05555e2df15771c6cbc64b5b08d31551cdaef5f4d499", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ec1a2b892c821d2638dd9ada48723f3ab543ccfc67c2a49ec79bf99ad33583e3", "generation_seconds_batch": 23.302728619077243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.31941294670105, "target_z": 0.792968839520206, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.27974069118499756, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/38cb3714acdb405c4e2a05555e2df15771c6cbc64b5b08d31551cdaef5f4d499.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "38e7b35a8ebb91b418b4f5139c4f573849426a080f2fe44122cc67da990223a8", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "25b60f5effb037d9a28520c326fc1c923cd23bb8d18530a6fb4693c2aa4c148b", "generation_seconds_batch": 17.752706032944843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.384434700012207, "target_z": 0.8710937679040412, "ce": 6.23706579208374, "genuineness": 4.595211505889893, "identity": 0.701968252658844, "hypothesis": "De deur 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_high/38e7b35a8ebb91b418b4f5139c4f573849426a080f2fe44122cc67da990223a8.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "390b4d432e1be6fc8332ad624642407f1c21981be7bda13ffed782009d78f66a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9a4f52f7b54559390f9feb4a39583469960c8182f598ae7987a70b3ffbebd8b3", "generation_seconds_batch": 22.740253194002435, "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": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.5636703968048096, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/390b4d432e1be6fc8332ad624642407f1c21981be7bda13ffed782009d78f66a.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "3910ca5a2fac437d3d5f2ecfe6662edaa72d101a99f245b1e0cdb896e218adb3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8b4278dc0c057d8024ce7b10de9c4b2d235b4f5bf2c86740f6768d948fd22023", "generation_seconds_batch": 24.97288221505005, "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": 5.982900142669678, "genuineness": 4.65797233581543, "identity": 0.8202853798866272, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3910ca5a2fac437d3d5f2ecfe6662edaa72d101a99f245b1e0cdb896e218adb3.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": "3954c09a0572b78fd8be8d7cf9fe9a6826ee4d4095264ec0938eb1dd0e06c999", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1b35f9a42d7d903a86eba438bdd07a5014635b4bb8c786670f8e8ee568c8d9e5", "generation_seconds_batch": 22.21040278195869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7401394844055176, "target_z": 2.5, "ce": 5.279973983764648, "genuineness": 3.8044304847717285, "identity": 0.5034545660018921, "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_high/3954c09a0572b78fd8be8d7cf9fe9a6826ee4d4095264ec0938eb1dd0e06c999.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "39c127226a21e8f413dc46583b56f0e30202cf37ae050ed954fb1275c3f2e605", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3f4ca958a69e6bf473c4d20ff7922d5d70344825588bee16014a9ab4990ff9d4", "generation_seconds_batch": 24.54003436199855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7648122310638428, "target_z": 1.3281249283838352, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.4517630338668823, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/39c127226a21e8f413dc46583b56f0e30202cf37ae050ed954fb1275c3f2e605.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "3a0b83145d2a0289dc78a64aa2ebb5a1d828fc7b75e7255f193b55378bdb7d0a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3484b0155aa47d7a4cd5f881d94ba04baa2ff27126862bfff426aff20acc694d", "generation_seconds_batch": 17.48535330593586, "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": 6.006728172302246, "genuineness": 4.20609712600708, "identity": 0.5374880433082581, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3a0b83145d2a0289dc78a64aa2ebb5a1d828fc7b75e7255f193b55378bdb7d0a.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "3ae39a80098e2627556d0bb5029351b1266f4d3d8aeb94bad6bc019ecdb508c3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "25027f225080d86946e74fee756d429f1d06501223f4c222a7486f52afaac72d", "generation_seconds_batch": 22.20982741599437, "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": 5.895530700683594, "genuineness": 4.218649387359619, "identity": 0.6559012532234192, "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_high/3ae39a80098e2627556d0bb5029351b1266f4d3d8aeb94bad6bc019ecdb508c3.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "3b4427eb23ee346787cde36ad06477490a6906ed91ac23a3df2d89117bcdd006", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9a89fffea90a685c007d69bf34fac4f7d81953bdb3e6e1fdcdfc8d043edfc931", "generation_seconds_batch": 18.09574145905208, "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": 6.475345611572266, "genuineness": 4.105680465698242, "identity": 0.4670792520046234, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3b4427eb23ee346787cde36ad06477490a6906ed91ac23a3df2d89117bcdd006.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "3b60bb15f0965ba3eb287029afda3b8bce01bec4a8b5795b0bc1c68e3d3cc1d3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "43793c965aad38ce8301b1423e35017be45c182fc8325eea1345f627aceb0851", "generation_seconds_batch": 22.31284346897155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.050908327102661, "target_z": 1.6718750716161648, "ce": 5.657250881195068, "genuineness": 4.2751336097717285, "identity": 0.7956066727638245, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3b60bb15f0965ba3eb287029afda3b8bce01bec4a8b5795b0bc1c68e3d3cc1d3.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "3b8c91df2f710381d48fbb69fb0b94cea59d2b7ce3c9ec4c3940c11258c1981e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c5ca1037fded0a7eb0c0293f211e3148b38d4ff8e366c0b1b20c200e2cdc52d5", "generation_seconds_batch": 22.661905245040543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5107040405273438, "target_z": -0.17871097778409273, "ce": 6.181467056274414, "genuineness": 4.808597087860107, "identity": 0.576432466506958, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3b8c91df2f710381d48fbb69fb0b94cea59d2b7ce3c9ec4c3940c11258c1981e.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "3c51b60af1208ea32a36afcded44b23c488ec590d00534a6c4302bd742690462", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9051dc60f31fafa5765c82f217b4ce0befed121f3154249a4016f92463951f68", "generation_seconds_batch": 22.48584399803076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7228376865386963, "target_z": 0.07617181233585575, "ce": 6.117925643920898, "genuineness": 4.607763767242432, "identity": 0.7582534551620483, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3c51b60af1208ea32a36afcded44b23c488ec590d00534a6c4302bd742690462.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "5426916e179e922ddbe6be8b73cc923e23a4b1ab0c1b3a23d355020b53d4b7d4", "generation_seconds_batch": 16.866215236950666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.351923704147339, "target_z": 0.832031160479794, "ce": 6.181467056274414, "genuineness": 4.250029563903809, "identity": 0.7278374433517456, "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_high/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "3c6e2a79f973bd333e440b9bdd00a0fa2b733c0d9de83d8b8095f3a93d8dbff5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "65f2719ca9cb145bfa7a3d0e2cf2d5418b1d854afdf0f53dc08d785ea33760f2", "generation_seconds_batch": 18.739524059928954, "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": 5.974957466125488, "genuineness": 4.557555198669434, "identity": 0.7186034321784973, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3c6e2a79f973bd333e440b9bdd00a0fa2b733c0d9de83d8b8095f3a93d8dbff5.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3c6e6c80f408d08ff82a6f4323c648b7c6ba14e0165a5bdb76b57600b5915ced", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ffab3f4b1fdec0f57361e65d201ee8c687c7a14404eb6da0246d45a806a165fc", "generation_seconds_batch": 22.750165203004144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8493406772613525, "target_z": 1.4296875358080825, "ce": 5.959072113037109, "genuineness": 4.231201171875, "identity": 0.7587963342666626, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3c6e6c80f408d08ff82a6f4323c648b7c6ba14e0165a5bdb76b57600b5915ced.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "3c9287083fba98ce999edee35de07e54e634493c0105550bb743e3b26b4ed1dc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2d78e06efe7dd1483300f7b1403f388383e9163df40e5f6ac3e5d02672a1aa36", "generation_seconds_batch": 22.89164546399843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.798425555229187, "target_z": 0.16699213826388667, "ce": 6.181467056274414, "genuineness": 4.18099308013916, "identity": 0.4339677691459656, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3c9287083fba98ce999edee35de07e54e634493c0105550bb743e3b26b4ed1dc.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "3cb9f67510039e38c6f7d141304306d78556300c189545d36b22193d76d58949", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "defd32ede8987cdf23f6f2e10c61a345dcbb17ba843b6e75cca340b49b4f4765", "generation_seconds_batch": 23.249505473067984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.23488450050354, "target_z": 0.6914062320959588, "ce": 6.252951145172119, "genuineness": 4.494794845581055, "identity": 0.7054044008255005, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3cb9f67510039e38c6f7d141304306d78556300c189545d36b22193d76d58949.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3cf265e32c018b20a4f473af00c49851d888a9804b078caf85f0b6a4da4fd8c5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "87590f776fe601a454637460e19361bba67a0c9b455ed6f9352b90ca1cf58aa9", "generation_seconds_batch": 21.5504004759714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.831170082092285, "target_z": 2.609375071616165, "ce": 5.069493293762207, "genuineness": 3.9299514293670654, "identity": 0.8237518072128296, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3cf265e32c018b20a4f473af00c49851d888a9804b078caf85f0b6a4da4fd8c5.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "3d201fb0b90cb3c6bcf6f498163b127888167b65d3bc2773fa318908cab31fa0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9c28cc0639af1d211f90f9da3e5d4c1ba12aeb26900996a5d38d5acb3c520586", "generation_seconds_batch": 22.101699391030706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.117845296859741, "target_z": 0.5507813037121236, "ce": 6.221180438995361, "genuineness": 4.174716949462891, "identity": 0.23669755458831787, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3d201fb0b90cb3c6bcf6f498163b127888167b65d3bc2773fa318908cab31fa0.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "3d29fe7b48c5bb3ac3af64c4e0fb0ca7bba5af8d787e0d45bc5a556eb0618c60", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fa80bfb7223096d0530c7316e6547b1038cf2a3674254729637ca9381754668e", "generation_seconds_batch": 22.707033994956873, "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": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.43160903453826904, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3d29fe7b48c5bb3ac3af64c4e0fb0ca7bba5af8d787e0d45bc5a556eb0618c60.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "3de3d371a23d2c1c1b66fb367103820da2d30a14ff02beab50434c9ae380a43c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0a1d99df63b1f98845d27af8df0f9fef729de727f849d6a69a36b8010703eddc", "generation_seconds_batch": 17.07032868207898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.901358127593994, "target_z": 1.4921875358080825, "ce": 6.260893821716309, "genuineness": 4.21237325668335, "identity": 0.5012978315353394, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3de3d371a23d2c1c1b66fb367103820da2d30a14ff02beab50434c9ae380a43c.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "3e6a151b271f1cc5f0f49950546bae91e7318bb9e0ba521df0293a6b3dffe14e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "18a8d2686226e9bc0aa2de866c7968478501cf3096efa6df19e99d7a555378e8", "generation_seconds_batch": 18.30825455102604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.66635000705719, "target_z": 0.00830074320391242, "ce": 6.23706579208374, "genuineness": 4.6956281661987305, "identity": 0.7257020473480225, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3e6a151b271f1cc5f0f49950546bae91e7318bb9e0ba521df0293a6b3dffe14e.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3ebbed9a17297b623530146bb8f5af7755f2ac03005ad60dd91c4201ca33a51a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "4f7da8a2b3206e21b723368c9e4669814da63753bdc5e561e5e450623ccd61ac", "generation_seconds_batch": 17.306138889980502, "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": 6.141753673553467, "genuineness": 4.808597087860107, "identity": 0.6124564409255981, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3ebbed9a17297b623530146bb8f5af7755f2ac03005ad60dd91c4201ca33a51a.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "3f343bd4e59a7f61a7b801fe03ad0c76218e35eb5bf819965889ac35c5b92f1b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c0ef5a9810dfc9c41b888cabeff1bea8965a4a698de6d3cfd74162d0865b0743", "generation_seconds_batch": 26.706156763015315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5293978452682495, "target_z": -0.15625, "ce": 6.372091293334961, "genuineness": 4.507347106933594, "identity": 0.32502299547195435, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3f343bd4e59a7f61a7b801fe03ad0c76218e35eb5bf819965889ac35c5b92f1b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "3f47b724f123cc75e4b15bf6af755a3b503029ff9133ca635ebc62537b3a3736", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "61a56f9c30788092c728c4f7d13605728f9782b4973b84d3f0be0ab5e638b831", "generation_seconds_batch": 22.229893729905598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.844174385070801, "target_z": 2.625, "ce": 5.824047088623047, "genuineness": 3.961331605911255, "identity": 0.4986175000667572, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3f47b724f123cc75e4b15bf6af755a3b503029ff9133ca635ebc62537b3a3736.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "3f5e9af2832e8cf37a384d3dd03f79b0871c257c9a44f1ec775da774035de9f0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "353cf24a70ec8b03ed281b1f32af3413df7c4ea0410620078c01171d5d9bf014", "generation_seconds_batch": 22.638067246996798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1438539028167725, "target_z": 0.582031160479794, "ce": 5.736677646636963, "genuineness": 4.419482707977295, "identity": 0.42981016635894775, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3f5e9af2832e8cf37a384d3dd03f79b0871c257c9a44f1ec775da774035de9f0.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "3f701fd45449a7171f27d84bd6c203ddf6bdc88027dcf8e93b23a30512a15ece", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "918309fa5f91c1d5ae4fa4f70b37f537354d9f1d30854f74ffdddda310db0911", "generation_seconds_batch": 24.495827788021415, "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": 6.38003396987915, "genuineness": 4.519899368286133, "identity": 0.6406885981559753, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3f701fd45449a7171f27d84bd6c203ddf6bdc88027dcf8e93b23a30512a15ece.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "3f9bfc4f3858dc465c685ce2c9bf421a858d162116a18f1878f4df9f4e44da64", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e8a6d7f547c04d04fceaaa00cf09e8ee559d5801fa30d9f0bc0ef42646437601", "generation_seconds_batch": 17.299062434001826, "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": 6.403861999511719, "genuineness": 4.444586753845215, "identity": 0.5458454489707947, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3f9bfc4f3858dc465c685ce2c9bf421a858d162116a18f1878f4df9f4e44da64.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "3fa8588e7376889aac94b5c5e1fd367125ea5a41c92c844a81a90e1f9e55348b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "58fdfcf3927f59254e83b797b70f7b3af3feac2eb3d1868cd19547984f71eaa2", "generation_seconds_batch": 17.269254012033343, "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": 6.054384231567383, "genuineness": 4.607763767242432, "identity": 0.7220121622085571, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/3fa8588e7376889aac94b5c5e1fd367125ea5a41c92c844a81a90e1f9e55348b.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": "40060dd2983dc8e585ff010cf7ec68068bb5f6997d9102b6110919cb47a04965", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a57e46e27599e56a885010f0c083e7e36c547ed26a6e22c5fcd7908bde32c97a", "generation_seconds_batch": 16.94114570110105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.309659481048584, "target_z": 0.7812498567676703, "ce": 6.205295085906982, "genuineness": 4.519899368286133, "identity": 0.44868820905685425, "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_high/40060dd2983dc8e585ff010cf7ec68068bb5f6997d9102b6110919cb47a04965.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4066fa657ac3d47851d1adc3ea37163613451d1914b8eb5eaf0bba5e6a1776c8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7961a5ebb34dc164f1988a9c0d61fa6b855b2a002e3478b7daced496901f493d", "generation_seconds_batch": 17.17835149099119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.206960678100586, "target_z": 1.8593750716161648, "ce": 5.847874641418457, "genuineness": 4.344170093536377, "identity": 0.7427924275398254, "hypothesis": "La porte 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_high/4066fa657ac3d47851d1adc3ea37163613451d1914b8eb5eaf0bba5e6a1776c8.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "40b0515490a70b23eb3f7127ab834b4f9c20377d04c896541e2ffdecaf205508", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f99a3dff43125064e1e8f7b421c0f71dfd0bbcaabbcf029eb595482560921be", "generation_seconds_batch": 16.290564085938968, "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": 4.7332844734191895, "identity": 0.6603835225105286, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/40b0515490a70b23eb3f7127ab834b4f9c20377d04c896541e2ffdecaf205508.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "40dd8895c94ead826d14fdf07444a427cf7ff7b08109bd46eeb9b35551bf23a3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "994a3120db14b59b725f567cc6a8430550779e8bee23e496b6d56699e48cde68", "generation_seconds_batch": 24.58434047398623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.5102738738059998, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/40dd8895c94ead826d14fdf07444a427cf7ff7b08109bd46eeb9b35551bf23a3.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "40e9060d641ef3ccdc433a83a0e518fc9c037232d84bd2532c6703ff4d99c294", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4cb99f12ef7e745e24225e0948d6f8923a58c310b17617d1bc95c3b716d23e9d", "generation_seconds_batch": 23.713507132953964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 6.149696350097656, "genuineness": 4.162164688110352, "identity": 0.5245971083641052, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/40e9060d641ef3ccdc433a83a0e518fc9c037232d84bd2532c6703ff4d99c294.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": "424a875647cdcca2dd1318d4af9896590117e714191beab6bd66aedc121bccbe", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ea8537ca0a6f5523629b798691fd816eca3ea312fb134c3199e73ce1e978c45", "generation_seconds_batch": 17.236942888004705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.712794780731201, "target_z": 1.2656249283838352, "ce": 6.078212261199951, "genuineness": 4.444586753845215, "identity": 0.7129333019256592, "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_high/424a875647cdcca2dd1318d4af9896590117e714191beab6bd66aedc121bccbe.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "42686bcda2ff24e81238b9f239230707e90c0ba4fe4f9ceb88a624b731ac69fd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d7deeab6659e1c66565e9a4117a956c9eaa49aa453a7ef95438a3acf28734ff0", "generation_seconds_batch": 23.341008462943137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.200458526611328, "target_z": 1.8515626074242473, "ce": 5.8875885009765625, "genuineness": 4.507347106933594, "identity": 0.7722399234771729, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/42686bcda2ff24e81238b9f239230707e90c0ba4fe4f9ceb88a624b731ac69fd.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "42fc3f542a8753f96bf4f9d9b6430dc9818c0b162bbbfaf927b2213d82adc627", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "efda36c6e443b36d8d2d313f1a9450f65969e64b5eab33afc4bd06f5b5299574", "generation_seconds_batch": 22.934322233079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.98722243309021, "target_z": 2.796875071616165, "ce": 5.283945083618164, "genuineness": 3.7291181087493896, "identity": 0.36016201972961426, "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_high/42fc3f542a8753f96bf4f9d9b6430dc9818c0b162bbbfaf927b2213d82adc627.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "430f5fb068d01460ac0af22ebcba3fb29f4a1dc28f0b6b15ce1f4468bedc0f9e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "62649748774263e85e76156b6129e53d5c1253d9e7660da624e097dcd55eb302", "generation_seconds_batch": 23.26225476397667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.712794780731201, "target_z": 1.2656249283838352, "ce": 6.260893821716309, "genuineness": 4.306513786315918, "identity": 0.45432427525520325, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/430f5fb068d01460ac0af22ebcba3fb29f4a1dc28f0b6b15ce1f4468bedc0f9e.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "ae11b238716118bdeb024a7c1b1ac432d1b465fe83d44f213b6e206a19efe3e4", "generation_seconds_batch": 17.23343314800877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.206960678100586, "target_z": 1.8593750716161648, "ce": 6.149696350097656, "genuineness": 4.15588903427124, "identity": 0.4953908324241638, "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_high/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "a2d623d2b4bc852082d98665aec3d5c621f931043d04ba69ad06e22bd3c053a5", "generation_seconds_batch": 16.865519181010313, "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": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.5512871742248535, "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_high/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "4472c75088f07ef37336fac87f7d291064905e0dcd9ecb2c7e5442ffaf357500", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e5b450c99abf64f633fb6ab1835848266cb3e7acdbf89114724c5e161f1af845", "generation_seconds_batch": 22.35609524208121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.792156934738159, "target_z": 2.5625, "ce": 5.839932441711426, "genuineness": 3.898571252822876, "identity": 0.19225946068763733, "hypothesis": "I found the little box. It was beside the old window. We can look inside Dick.", "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_high/4472c75088f07ef37336fac87f7d291064905e0dcd9ecb2c7e5442ffaf357500.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "44934da3ef3d87bde33ff73f387a042b28c09f4170cab4dd44fc5f2da75685cc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3273043d3aafd707dfb5e6b8f36e21e07472ff07b51c922b18f42acfdf3c97c3", "generation_seconds_batch": 23.229214551043697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.709426999092102, "target_z": 0.06005860494002576, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.49311384558677673, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/44934da3ef3d87bde33ff73f387a042b28c09f4170cab4dd44fc5f2da75685cc.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "451869f73dcb71e24748dcca7f0eb14011165f07780813b38700ca886ed9c115", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d412353a8a376bac9445369af82cd63abab3f8d575b7095c53ccbb2a5afbc4de", "generation_seconds_batch": 23.92921092105098, "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": 5.95112943649292, "genuineness": 4.362998008728027, "identity": 0.5801262855529785, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/451869f73dcb71e24748dcca7f0eb14011165f07780813b38700ca886ed9c115.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "45b45a036bc5181234bb073ec7342b60eaed72f06830c73daa371b7b85eba596", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4a01b46a62b2e6d7c3ee735a29f8af42fddae837ff6ba38fedf9289c7bd36920", "generation_seconds_batch": 22.31284346897155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 5.550024509429932, "genuineness": 4.300237655639648, "identity": 0.8082668781280518, "hypothesis": "La portée. La portée est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/45b45a036bc5181234bb073ec7342b60eaed72f06830c73daa371b7b85eba596.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "3f49f6babe93db44eeaa54480ed7f206e92790c9142a150fc97dcf986f80e739", "generation_seconds_batch": 18.015127991093323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.725799083709717, "target_z": 1.2812498567676702, "ce": 6.340320587158203, "genuineness": 4.344170093536377, "identity": 0.467068612575531, "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_high/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "46117b0c86118b6912ec924e57fec278aaecd98246632aad9c2c124cac44d796", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ed3728c0808436d0d68fdd3fa9a70fe774fc766260eed48cc882fc9f3677fe1c", "generation_seconds_batch": 23.112488563987426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.381183624267578, "target_z": 0.8671875358080824, "ce": 6.372091293334961, "genuineness": 4.231201171875, "identity": 0.3950542211532593, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/46117b0c86118b6912ec924e57fec278aaecd98246632aad9c2c124cac44d796.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "463b523922a26c6da1727d35dd2f1e0d0815849994effd89b5b22e48ef04da34", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4dc8d279012307239821ffef7925b320b9d4a96e2710acca56870219d47bb601", "generation_seconds_batch": 22.33074641006533, "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": 6.181467056274414, "genuineness": 4.231201171875, "identity": 0.5077080726623535, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/463b523922a26c6da1727d35dd2f1e0d0815849994effd89b5b22e48ef04da34.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "465e41997a8fdac222f92e9417308c9eb0acb3a7448476bda5e2da383e99b0ab", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "552dd7cc4498bb1b66e6ad2d57c010b39f886b7f82d008f2fe8157f4a3ab7a18", "generation_seconds_batch": 22.37107196706347, "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": 6.181467056274414, "genuineness": 4.306513786315918, "identity": 0.5174920558929443, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/465e41997a8fdac222f92e9417308c9eb0acb3a7448476bda5e2da383e99b0ab.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "46bf4097e267bf449fead58382fc678eaa272111304a33810a2f31612a172edc", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "84c64a9a547f45dd99fb96f16079f089ea663932fb066a5bfb0accae198d2a2f", "generation_seconds_batch": 17.4862761850236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8902689218521118, "target_z": 0.2773437679040412, "ce": 5.784333229064941, "genuineness": 4.356721878051758, "identity": 0.5104885101318359, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/46bf4097e267bf449fead58382fc678eaa272111304a33810a2f31612a172edc.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "46ccc79f83f278dac025fa72398d4bbd5052bed82f3752548e1c7ee8c0f3b9d8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8ad10194e8a598418f94fc45863e4fb551a06f510014c2a127980ba39eb92fc5", "generation_seconds_batch": 23.82093454990536, "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": 6.141753673553467, "genuineness": 4.7458367347717285, "identity": 0.6301218867301941, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/46ccc79f83f278dac025fa72398d4bbd5052bed82f3752548e1c7ee8c0f3b9d8.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "46e4450280edbd0b0bd704f3e439459d25641c4a3f1f2475717b9dae8d9f6423", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f6d5af61230b367def86c71bb094c94880290535ca64c28dc57bf8958dad70a0", "generation_seconds_batch": 23.85429213603493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.316161632537842, "target_z": 0.7890623209595878, "ce": 6.268836498260498, "genuineness": 4.7332844734191895, "identity": 0.8110264539718628, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/46e4450280edbd0b0bd704f3e439459d25641c4a3f1f2475717b9dae8d9f6423.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "471c40d4eecbbb3e18b0fd5339f4ef22afdb576a6d0a3ff45ad09996b70e5b22", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "936ed29bfcc79c8d63cd8c30e18273ea0e64c52efe70ce8f1119059d7e9774a7", "generation_seconds_batch": 25.53896590997465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.862344980239868, "target_z": 1.4453124641919175, "ce": 5.184661865234375, "genuineness": 4.005263805389404, "identity": 0.35088393092155457, "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_high/471c40d4eecbbb3e18b0fd5339f4ef22afdb576a6d0a3ff45ad09996b70e5b22.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": "47321edee103a36c8bc6806f1cc9095250c5d5ee762455219c53d131b1c94760", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9c9545219b6eb91202aa0018d752c90e6e904605af8824f5d33721adcd32b1e", "generation_seconds_batch": 22.17140672390815, "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": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.7138538360595703, "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_high/47321edee103a36c8bc6806f1cc9095250c5d5ee762455219c53d131b1c94760.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4751f7ed71d5757f67369245a9626687ec55478704bac15eb630c0ffb8a0da3b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3b893a2f61ec62ef72e1f5ebc26d5d8fe4099d2b148edb67439c10027995ea4c", "generation_seconds_batch": 18.31541846797336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.025189161300659, "target_z": 0.4394531160479794, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.5586118102073669, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4751f7ed71d5757f67369245a9626687ec55478704bac15eb630c0ffb8a0da3b.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1bd5156693f95a7510d4c5feba600be3a859b8a7e636679094e600e60a1f3b1d", "generation_seconds_batch": 16.429206655011512, "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": 6.038498878479004, "genuineness": 4.432034492492676, "identity": 0.7507513165473938, "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_high/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "47f85ed3a5f616b0e18abfd32c9ace842fb84ebf622af3fdfda7ab8b1d7a6458", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "82c2c489d8d91a2733a73026e5d54a1eac2c3c274836ac9e2f8f429a399f1f1f", "generation_seconds_batch": 22.859268696047366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.688122034072876, "target_z": 2.4375, "ce": 6.125868320465088, "genuineness": 3.9676074981689453, "identity": 0.6077251434326172, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/47f85ed3a5f616b0e18abfd32c9ace842fb84ebf622af3fdfda7ab8b1d7a6458.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4808645f04bdc93820f35da950f2264b470475b4f83cc093e180d87b2fd8be51", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "39f7e9d871ed8a72cb55ff8d2dac759af7a305f0ffbb6f587d56ac20506adc7a", "generation_seconds_batch": 23.068849861971103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2199649810791016, "target_z": 1.875, "ce": 5.144948482513428, "genuineness": 4.168440818786621, "identity": 0.577738344669342, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4808645f04bdc93820f35da950f2264b470475b4f83cc093e180d87b2fd8be51.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "48367d848da90113330c8f64a364d405863f37a9158063a2801b8d7c016e5405", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e117b393006b21247b3eeb7c09d9e77fc78395c2bf4599d9e8acd122a047db53", "generation_seconds_batch": 17.290536138927564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0219380855560303, "target_z": 0.4355468839520206, "ce": 6.062326908111572, "genuineness": 4.394378662109375, "identity": 0.4754013419151306, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/48367d848da90113330c8f64a364d405863f37a9158063a2801b8d7c016e5405.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "483f91b5e69f27aca9db36db0cbe6776a2c3dff38f3d9725c6c5e7c32e5dde65", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "53907ac1a617f32eabd835f6e429b058b638a3e767e1d8a918ba868c680bc156", "generation_seconds_batch": 17.793326555052772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.013810396194458, "target_z": 0.42578130371212364, "ce": 6.284721851348877, "genuineness": 3.9676074981689453, "identity": 0.31391096115112305, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/483f91b5e69f27aca9db36db0cbe6776a2c3dff38f3d9725c6c5e7c32e5dde65.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "487668f962a2cb36fe5d3fa4985be8fb8176ec56824b83ef533f8043975df043", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "138bfdc0d2f32876725fba1c1d872b76cd25f37f86919864a6ba2848eeaba0cb", "generation_seconds_batch": 23.11722616502084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2283823490142822, "target_z": 0.6835937679040412, "ce": 5.744620323181152, "genuineness": 4.419482707977295, "identity": 0.5683977007865906, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/487668f962a2cb36fe5d3fa4985be8fb8176ec56824b83ef533f8043975df043.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "48ca4dadb7bff0ff5a0079c9aefddfb333d88313b878d8ca09d61862c307e9d9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9663c8a3cabd351b20f0fcfd785326d98c21ce59371100ccf1f04a5419fd110d", "generation_seconds_batch": 22.683026877930388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2654802799224854, "target_z": 1.9296875358080825, "ce": 6.221180438995361, "genuineness": 4.293961524963379, "identity": 0.788027286529541, "hypothesis": "J'ai trouvé la petite poite. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/48ca4dadb7bff0ff5a0079c9aefddfb333d88313b878d8ca09d61862c307e9d9.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "49448980d3acb4e519cfef354d2e788d16704fcd863a1b4fd1712338bcf07057", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4c895eebca7a2de19f61f0557e993fd5a90d99eec1f1ef12d28ae4a27389bced", "generation_seconds_batch": 17.32587613002397, "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": 6.117925643920898, "genuineness": 4.720732688903809, "identity": 0.6552377343177795, "hypothesis": "La porte 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_high/49448980d3acb4e519cfef354d2e788d16704fcd863a1b4fd1712338bcf07057.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "49497c1d62eb73c311c7099234e096876d50b4c49e56ea1ece44a1c1c4533398", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "32a492099ba1a56c108240601284a6e33dcd2042f3a3dd8276c5175fc9be2688", "generation_seconds_batch": 23.063927563955076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.222636699676514, "target_z": 4.28124985676767, "ce": 5.9431867599487305, "genuineness": 3.4561104774475098, "identity": 0.4974473714828491, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/49497c1d62eb73c311c7099234e096876d50b4c49e56ea1ece44a1c1c4533398.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4965cbe3ee4e7e5f170d2e27da3e1ebbd398d7394e344756bc53d0a3dc7d8d6a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c8b4f698f80b0ea19e900c66747d77ce2ba06d762d140bc187283deba8d1c25a", "generation_seconds_batch": 17.829717296990566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532069683074951, "target_z": 2.25, "ce": 5.863759994506836, "genuineness": 4.250029563903809, "identity": 0.6959528923034668, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4965cbe3ee4e7e5f170d2e27da3e1ebbd398d7394e344756bc53d0a3dc7d8d6a.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": "49be9032d1592781ab8c887c8036b732a4fd1e056a4ee13a281af5339c65cbca", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7dbb0beb8c3b734b7a180e17ea9e5ae056a0495e35963a25e0eb36db01e2566e", "generation_seconds_batch": 22.17140672390815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1438539028167725, "target_z": 0.582031160479794, "ce": 5.990842819213867, "genuineness": 4.444586753845215, "identity": 0.5799069404602051, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/49be9032d1592781ab8c887c8036b732a4fd1e056a4ee13a281af5339c65cbca.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "ef249226c2ad3821ade9c29627c9f1eeb0ac6b77fdc96bb80c64ddd6f22bd331", "generation_seconds_batch": 16.76059895195067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5502402782440186, "target_z": 1.0703124641919175, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.6607146263122559, "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_high/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "4a0610a453e178de44792388d6c06954497868bcccdcbd1b8445d03398f28ee3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "470916fcb4be344a3660bf219abf8e8e2e5a583dc6720b345183a59dbe63c964", "generation_seconds_batch": 24.214766227989458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823331832885742, "target_z": 1.3984373925757527, "ce": 5.8875885009765625, "genuineness": 4.293961524963379, "identity": 0.44472670555114746, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4a0610a453e178de44792388d6c06954497868bcccdcbd1b8445d03398f28ee3.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4a17376acb4228694e06f76b4a9221a16a7cbafbb9b4126852e14579dfb5b2c1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cde3441fb36f71e4384df149e7c56abbf07d60e30f488a4afb54525fc3fac0ef", "generation_seconds_batch": 22.448595992987975, "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": 6.133810997009277, "genuineness": 4.557555198669434, "identity": 0.5646781921386719, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4a17376acb4228694e06f76b4a9221a16a7cbafbb9b4126852e14579dfb5b2c1.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4aef037cc5b085b4df77e2cb85f77c7be241eefbd9fb94228d8a346dab0134f1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "49b2892b2473485726f1becfea15c3706b2936e14c9df49c0ca2b15142865580", "generation_seconds_batch": 22.005086333025247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9352049827575684, "target_z": 2.734375071616165, "ce": 5.164804935455322, "genuineness": 3.9173994064331055, "identity": 0.7141472101211548, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4aef037cc5b085b4df77e2cb85f77c7be241eefbd9fb94228d8a346dab0134f1.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "4b8d6c94f6c9d707cae5c24fb4807f5ef917a8dfeef6a75d73c21fad417a7851", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "241c05ea95d899b9799aa2b9b831ed44c07ea703eb0f56daf3251209b6c40782", "generation_seconds_batch": 22.700801557977684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6681787967681885, "target_z": 0.010498070374054093, "ce": 6.3641486167907715, "genuineness": 4.362998008728027, "identity": 0.49547335505485535, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4b8d6c94f6c9d707cae5c24fb4807f5ef917a8dfeef6a75d73c21fad417a7851.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "4c0378f01e5fae39700e8e71be43dfbe16d2246dc43451a37bb22ca1347ce9b7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "eb3ae47aef140bf9e2e71d75a794d252cf68c2e4a7f58eec4246ee9692d331fe", "generation_seconds_batch": 24.210008699912578, "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": 5.998785495758057, "genuineness": 4.15588903427124, "identity": 0.5898758769035339, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4c0378f01e5fae39700e8e71be43dfbe16d2246dc43451a37bb22ca1347ce9b7.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4c60a6e187845ce29185fb7847d66687c843a0f833d731f49d11ff065cff0bfa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f4abb87c5a1607244c21bcb8dc6c666eca16aafa0a75f6d3d08acb283043c85", "generation_seconds_batch": 21.81757136201486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.907860279083252, "target_z": 1.5, "ce": 5.855817794799805, "genuineness": 4.507347106933594, "identity": 0.7515584230422974, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4c60a6e187845ce29185fb7847d66687c843a0f833d731f49d11ff065cff0bfa.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "4c8beadc3f66273cf1704a3587e733540465148d2f4e265f6c7d4a6c70c85ac8", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b700f9830a9f2c14c783976f3004ddf6fa009ea65ffc0cd0b87f359ab8e5d18d", "generation_seconds_batch": 16.627182731986977, "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": 6.260893821716309, "genuineness": 4.607763767242432, "identity": 0.608945369720459, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4c8beadc3f66273cf1704a3587e733540465148d2f4e265f6c7d4a6c70c85ac8.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "4cea2ce275583495395a390ec3c17810e58ab14c7beee75105021ce8223cd306", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3084efa849be8426b5fe4e10cd054a426b0ac8a31cb9a1ff8b8a116a653d4802", "generation_seconds_batch": 22.58199268206954, "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": 6.2767791748046875, "genuineness": 4.469690799713135, "identity": 0.6444169282913208, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4cea2ce275583495395a390ec3c17810e58ab14c7beee75105021ce8223cd306.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "4d111a603f6e581d590b48c580c460606b530004cd4627187e02a07afc89609b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "412875d702a20ecd092ca8137bb2dec09fafeb2d2ce52b59a1ec448d8a34cc70", "generation_seconds_batch": 22.740253194002435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0008058547973633, "target_z": 0.4101560888636291, "ce": 6.1894097328186035, "genuineness": 4.494794845581055, "identity": 0.4815825819969177, "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_high/4d111a603f6e581d590b48c580c460606b530004cd4627187e02a07afc89609b.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": "4d8e25c56bd14e47b227cfc779f47f3894f815bb084f0dc05d95a9126b423622", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af6e7ffeb4e9398b03196aebc0303a969eeaed155aa1c0da88f68a30015cef32", "generation_seconds_batch": 24.333027897053398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7011263370513916, "target_z": 2.453124928383835, "ce": 6.213237762451172, "genuineness": 3.911123275756836, "identity": 0.5662038326263428, "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_high/4d8e25c56bd14e47b227cfc779f47f3894f815bb084f0dc05d95a9126b423622.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4e503735c66ede5018d5dd2d24b8557cab9a51e5b529b8ffdcb93600b62466d0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f323ae22af7fd56b3121019040b8f2ad8c4a604d083aa530be45a5c31c89896e", "generation_seconds_batch": 17.829717296990566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532069683074951, "target_z": 2.25, "ce": 6.070269584655762, "genuineness": 4.331617832183838, "identity": 0.6120332479476929, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4e503735c66ede5018d5dd2d24b8557cab9a51e5b529b8ffdcb93600b62466d0.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "4e84d32da83f57e4a219dd5bbef05f71932faece153981f8f55e12a3605ce16d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4d207a3dc200a7d518d30ffcd038fc2b6349700c382fab4b6c08dfcd0e633e38", "generation_seconds_batch": 16.050832613022067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3974390029907227, "target_z": 0.8867186962878764, "ce": 6.300607204437256, "genuineness": 4.645420074462891, "identity": 0.6331400871276855, "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_high/4e84d32da83f57e4a219dd5bbef05f71932faece153981f8f55e12a3605ce16d.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "4ea5536a648a49c66faf662d038d703d8c776e64240e28d7f684d8ef6487c4e8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "db2d2f0991c060cfbbdedc1b4c1751a17f1fdedbda91191269e72be87fdee55c", "generation_seconds_batch": 17.71516993304249, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.169862747192383, "target_z": 0.6132813037121236, "ce": 5.935244083404541, "genuineness": 4.7834930419921875, "identity": 0.6629831194877625, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4ea5536a648a49c66faf662d038d703d8c776e64240e28d7f684d8ef6487c4e8.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "4ea994cca71e43a80435e7e777ed040bdefe2c7f90b6371852b641c7162cb6e3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fab9e337e37deff8b9d148d61455b56307c31bff1071789bf50b11fe523f5efe", "generation_seconds_batch": 17.57903288397938, "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": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.700629711151123, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4ea994cca71e43a80435e7e777ed040bdefe2c7f90b6371852b641c7162cb6e3.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "4f36a04a55ecbe15b92f5f8623dfd06a05bf9c547519a28013348cdf0cdb576f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4747f3e29e252384e85af78d7349d95eadb7b16cd5895cd779a0074c9a61ba13", "generation_seconds_batch": 22.32308379805181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9048988819122314, "target_z": 0.29492195556818546, "ce": 6.046441555023193, "genuineness": 4.457139015197754, "identity": 0.7427103519439697, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4f36a04a55ecbe15b92f5f8623dfd06a05bf9c547519a28013348cdf0cdb576f.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "4f4544e0bc187961bce5394ef75b82ceccfb8b0fff384c2d825a31e08bfadba1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "37ec11fa617665e8766f36d600b471d3e37e15a9a5199f03529f016ce6be9e92", "generation_seconds_batch": 21.365153088001534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.303157329559326, "target_z": 0.7734373925757527, "ce": 6.1020402908325195, "genuineness": 4.632867813110352, "identity": 0.5899995565414429, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4f4544e0bc187961bce5394ef75b82ceccfb8b0fff384c2d825a31e08bfadba1.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "4f5da2790541421f4f76b5827f80ec7f4d10bca8d6e8dfd2e16514c04da44a1f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2dc89fa92f19662b1598a915a4e04fe067974974217561d7663818e2d8252844", "generation_seconds_batch": 22.003199038095772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6431859731674194, "target_z": -0.01953130371212364, "ce": 6.09409761428833, "genuineness": 4.720732688903809, "identity": 0.7830122709274292, "hypothesis": "J'ai trouvé la la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4f5da2790541421f4f76b5827f80ec7f4d10bca8d6e8dfd2e16514c04da44a1f.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "4fa9f742feb097187598e9451f0ac34abe709ef8f963ea0caec891178e236ddb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "58f6b578c7c59f3c28e9fb53f6aa667d142711a83ac0e60f3b398747df3ef6b0", "generation_seconds_batch": 16.948556926916353, "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": 6.2767791748046875, "genuineness": 4.833701133728027, "identity": 0.5540152788162231, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4fa9f742feb097187598e9451f0ac34abe709ef8f963ea0caec891178e236ddb.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "4fc3e04031f9f150739e003c8e6489a864ee9e3a2cbdc597957904e88e6980bd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "c21f966c8bc4d003d65ec855a929117862cbb186ab491e979ba358ada58b8b3d", "generation_seconds_batch": 22.10518890491221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.758310079574585, "target_z": 1.3203124641919175, "ce": 5.748591423034668, "genuineness": 3.8420867919921875, "identity": 0.6983067393302917, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4fc3e04031f9f150739e003c8e6489a864ee9e3a2cbdc597957904e88e6980bd.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "4fe1d42b527e09ac53ad719b0ee471ad918d9e86aca958e56de234fe027ef084", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "0b8e3a92d021c5a07d90fa88f6a1b293d9027bd3b11718bc7d12c240eddfc94b", "generation_seconds_batch": 23.19902868999634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.073955535888672, "target_z": 0.4980468839520206, "ce": 5.95112943649292, "genuineness": 4.482243061065674, "identity": 0.6140807271003723, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/4fe1d42b527e09ac53ad719b0ee471ad918d9e86aca958e56de234fe027ef084.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": "502d037d47080b1e637ed3decbe0cafe5d0c2b8ec824c496dd7c8d19317040d3", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3aa8b35648f72a341129ede4543817ab5af99fcaff73253d08972276c419aeb", "generation_seconds_batch": 22.34890999691561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.011895179748535, "target_z": 1.625, "ce": 6.157639026641846, "genuineness": 4.231201171875, "identity": 0.5343626737594604, "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_high/502d037d47080b1e637ed3decbe0cafe5d0c2b8ec824c496dd7c8d19317040d3.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "50d57b3aa77741463126d03e2a1455891fb5887f1d26521a117bda7f1029e8f3", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5efa6351f70299e2bd49a82b15159695cc5992bb53d5c70e08801f8a660b6e59", "generation_seconds_batch": 18.166802349034697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7084109783172607, "target_z": 0.058837835793873784, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.2605556845664978, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/50d57b3aa77741463126d03e2a1455891fb5887f1d26521a117bda7f1029e8f3.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "51150516b845152bf6cc8eac63ebb39a9b5090a9a7a00e5c789cf184b334ecc6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8d941b1e7d41bb5481beb4d5a9cb2402e2a71f7ca0255e9d832be96b3a2da3a0", "generation_seconds_batch": 17.012744573061354, "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": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.748073160648346, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/51150516b845152bf6cc8eac63ebb39a9b5090a9a7a00e5c789cf184b334ecc6.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": "51c06db04681302e26d054b563b9a59c2e3dfb0bd0053e466082ba216142f339", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c383f74db7094d2403b9219db6c8401647733341985c6f89ff50eba1531c1d5", "generation_seconds_batch": 23.850330410990864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9273667335510254, "target_z": 1.5234373925757527, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.41197991371154785, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de laag. 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_high/51c06db04681302e26d054b563b9a59c2e3dfb0bd0053e466082ba216142f339.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "51e2bcd302ebf2e14a1fa4350fd19a165773d367c29e7fd832bea90a2b452211", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1b0c6456accc9718aa969d1fa3dcee70f3339a930385465900b352c9b096d46d", "generation_seconds_batch": 17.259381529060192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9390352964401245, "target_z": 0.33593753580808244, "ce": 6.395919322967529, "genuineness": 4.419482707977295, "identity": 0.6762081384658813, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/51e2bcd302ebf2e14a1fa4350fd19a165773d367c29e7fd832bea90a2b452211.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "520f0e4926b3b3a15f8faec82d0e021846bdb14a4efd2b53f84488014229429f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "07f9886cd890dbbf340d45708d63dab9089b1be88305e714b22d706f86e69b8e", "generation_seconds_batch": 24.23062640801072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.751807928085327, "target_z": 1.3125, "ce": 5.974957466125488, "genuineness": 4.331617832183838, "identity": 0.8173471689224243, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/520f0e4926b3b3a15f8faec82d0e021846bdb14a4efd2b53f84488014229429f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "5261bba6a1194f281ff37f81e71032c0fdeb9d4177b2f2a67ef0fbd8429c607a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f8e4348c7505e46a7852114527fa9ee9db24aef5bec5c8d97696e76a74f528d5", "generation_seconds_batch": 22.89550835103728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.477380394935608, "target_z": -0.21875, "ce": 5.990842819213867, "genuineness": 4.419482707977295, "identity": 0.6105165481567383, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5261bba6a1194f281ff37f81e71032c0fdeb9d4177b2f2a67ef0fbd8429c607a.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "52a237ebaac0d43bfe719168a08ae7645f64667b03320b7b9be38aa63b746dfc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1a2bd8b39d19624a5abce5499df666b0322bac4396a6bd2065068375dae18245", "generation_seconds_batch": 22.813404069980606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.01323127746582, "target_z": 2.8281252148484946, "ce": 6.24500846862793, "genuineness": 3.9927117824554443, "identity": 0.48819592595100403, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/52a237ebaac0d43bfe719168a08ae7645f64667b03320b7b9be38aa63b746dfc.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "52a425d44bd2ff066987d8860a810c3c6980f5e1e3c90710129fb02aec965511", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e8ebc40fb296d2fa438e489a1f60685b707988a4f569f861ff2f454601955451", "generation_seconds_batch": 24.000451322994195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.805161237716675, "target_z": 2.578124928383835, "ce": 5.7406487464904785, "genuineness": 3.9927117824554443, "identity": 0.6761382222175598, "hypothesis": "He encontrado. La cajita estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/52a425d44bd2ff066987d8860a810c3c6980f5e1e3c90710129fb02aec965511.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "5362e79917161858cfcf269c1c61160e3d2e0fad511cf1528d165fba68d0bbe5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "18ec2841c1ab20d1e91a7b5ac2ace1ab04a663a5ff714c515a7dd4ce2386b1bd", "generation_seconds_batch": 16.861858085962012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.202373504638672, "target_z": 0.6523436246717115, "ce": 6.173524379730225, "genuineness": 4.65797233581543, "identity": 0.49851495027542114, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5362e79917161858cfcf269c1c61160e3d2e0fad511cf1528d165fba68d0bbe5.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "5380dc35231d5a7a145d0b9bb1abdfa0846a602ff752f408ead96a24025324ea", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e2ad063c139a18f91ab9979fd7e91fbee752e1d22cf4b70d3348b0e2a8a4bbd4", "generation_seconds_batch": 23.252391926944256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.384434700012207, "target_z": 0.8710937679040412, "ce": 5.190618991851807, "genuineness": 4.469690799713135, "identity": 0.5996177196502686, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5380dc35231d5a7a145d0b9bb1abdfa0846a602ff752f408ead96a24025324ea.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "53ee85e394350d64dc181fe6c7c84e666852103a9682ce753a6b069dc1bb6743", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "38d0abc4ad524c70c8fa337969a1cd4ab9adca1bbd35d33455f65d7af34df775", "generation_seconds_batch": 17.971125644049607, "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": 6.427689552307129, "genuineness": 4.457139015197754, "identity": 0.6314919590950012, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/53ee85e394350d64dc181fe6c7c84e666852103a9682ce753a6b069dc1bb6743.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "542112eb84d158b6db1ca211b7998505e51fb3574ababf9c590f9efdf4e32187", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d810b8fbde77c956a1c123d0b007dba412300126ae87669f21cd9e5ae8ceaf7b", "generation_seconds_batch": 25.05968925799243, "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": 5.998785495758057, "genuineness": 4.2751336097717285, "identity": 0.7953220009803772, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/542112eb84d158b6db1ca211b7998505e51fb3574ababf9c590f9efdf4e32187.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "544e2667ba5199722d353c61b84fd9fd560c314e38a11274c36c272de43eb73f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3686d0bbbc5dbe708e6cf8b32ce0fdd180b0d630ed8afbaa19f6d385d9c3b643", "generation_seconds_batch": 23.666435455088504, "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": 5.95112943649292, "genuineness": 4.7458367347717285, "identity": 0.7018606662750244, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/544e2667ba5199722d353c61b84fd9fd560c314e38a11274c36c272de43eb73f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "54574b810106ec920eac4f847e90c80f8634bc68c0de509b133ee1b9df3521a4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "193b36ebf5828b9a4bea4e8af84dd674a222b27b2a3d9dd6a160a72a15b3c8fa", "generation_seconds_batch": 22.108564383001067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7606316804885864, "target_z": 0.12158204691603607, "ce": 6.372091293334961, "genuineness": 4.306513786315918, "identity": 0.6744973659515381, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/54574b810106ec920eac4f847e90c80f8634bc68c0de509b133ee1b9df3521a4.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "4e691135b54b879b7ba5824c9f30731bbf78bd94a5912ab7a311c9e12846930f", "generation_seconds_batch": 18.015127991093323, "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": 6.300607204437256, "genuineness": 3.9550557136535645, "identity": 0.6105207204818726, "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_high/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "54988cbda4238d7b64851e6825f3c8a4748be6662421a287e982d5dc6c388874", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ee0a12fe9ef54ea9bfada2804258301f26605df83aef8cd603df508869f0203a", "generation_seconds_batch": 22.494268746930175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8358131647109985, "target_z": 0.21191409383207213, "ce": 6.109982967376709, "genuineness": 4.256305694580078, "identity": 0.3911755084991455, "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_high/54988cbda4238d7b64851e6825f3c8a4748be6662421a287e982d5dc6c388874.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "54a7138fbab245455b09717abf102b139aa3c330145cd5f5a4a42f2479df601e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "df45eb2197b8b3c51582771854df666fa151381f969cbf60a22f8093486a3c77", "generation_seconds_batch": 16.242419044021517, "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": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.47565293312072754, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/54a7138fbab245455b09717abf102b139aa3c330145cd5f5a4a42f2479df601e.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "54ba2043b7348a14bbe94ed523a1683ade333d112658263ef78ba36312f80e67", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1ad02181887a59a016f2066c84fb69b8ef172e41ece6286ffcfa78e86bc05854", "generation_seconds_batch": 22.838779081008397, "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": 5.013894557952881, "genuineness": 4.005263805389404, "identity": 0.41622817516326904, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/54ba2043b7348a14bbe94ed523a1683ade333d112658263ef78ba36312f80e67.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "550e616e91b2382efc33b48de54edf8ca0dbc9d9783f05f97351e0285bc48349", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a870c9b6e2736b4815a245f4b6c5c7afc2d39ebf3acaec96685676309379dc40", "generation_seconds_batch": 17.47234323504381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7297463417053223, "target_z": 0.08447269877209788, "ce": 6.149696350097656, "genuineness": 4.557555198669434, "identity": 0.5464496612548828, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/550e616e91b2382efc33b48de54edf8ca0dbc9d9783f05f97351e0285bc48349.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "5540f9d9029f337cd42e2f09efd32a52eb6f935d119b5196167815a25cd9376b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "faf20dc1d05791e9637d0e10da13e7e94751e6657f1e66cd0e815e8491e3c927", "generation_seconds_batch": 20.80498196708504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.666908860206604, "target_z": 0.008972216365611407, "ce": 6.348263263702393, "genuineness": 4.507347106933594, "identity": 0.337284117937088, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5540f9d9029f337cd42e2f09efd32a52eb6f935d119b5196167815a25cd9376b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "559801ef9355f01617eed4f3d9804703a30da6e7e0d0edb312dae2af00d7f4e3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c77ec1a794a8a3ba324b906683fcdde4293af9eef8eeb87d7b2a0807ba50577d", "generation_seconds_batch": 22.97565673198551, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6025474071502686, "target_z": -0.06835934814393818, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.806334912776947, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/559801ef9355f01617eed4f3d9804703a30da6e7e0d0edb312dae2af00d7f4e3.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "55cb771bdb56e7996378ebb4fef0d602f3c4367d0f1998a143ab67cbcfb83fdf", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "abb285a83d86269c246adea2e794a64011b47f99a0c180e4ce2ef95b29f12583", "generation_seconds_batch": 23.246912455069833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.390936851501465, "target_z": 0.8789062320959588, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.2995232343673706, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/55cb771bdb56e7996378ebb4fef0d602f3c4367d0f1998a143ab67cbcfb83fdf.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "55d6c54be03cbbf6c64fd344e744821f9a846e3bf2f4bb4834b71982491b8f0f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "01b784abcae21dd03bcea8e3c56f11265aa017536839f1ea82d107b20d68a01f", "generation_seconds_batch": 17.053619588026777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9663798809051514, "target_z": 1.5703124641919175, "ce": 6.0146708488464355, "genuineness": 4.3504462242126465, "identity": 0.46456265449523926, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/55d6c54be03cbbf6c64fd344e744821f9a846e3bf2f4bb4834b71982491b8f0f.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": "562bca46b4528d682022099a5ba684df3d483af3e7d5ab7107af4389b33bc5b6", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59ab64a67b3c45d895b1b0972626076e6abaf970926a6f2a177c4aa7ba9955f4", "generation_seconds_batch": 22.045719899004325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.200458526611328, "target_z": 1.8515626074242473, "ce": 5.2640886306762695, "genuineness": 4.093128204345703, "identity": 0.5600020885467529, "hypothesis": "Deine Nachricht kam, als sie schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, woran wir reden können.", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/562bca46b4528d682022099a5ba684df3d483af3e7d5ab7107af4389b33bc5b6.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "5674c20de6962834dd331481a44f63fa287a434493ad91f0390fc42586fab2eb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3a590b5a5e74926b3eec9f29eec5990cf65bc483e282d841c6192a8d400f3428", "generation_seconds_batch": 23.99544641096145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.611432075500488, "target_z": 3.546875358080824, "ce": 5.760505676269531, "genuineness": 3.7165660858154297, "identity": 0.30659058690071106, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5674c20de6962834dd331481a44f63fa287a434493ad91f0390fc42586fab2eb.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": "5695aa3b8f891e87d79e60dee8274afef1fb0832a5eb238059601066db17e7c8", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acfbd6e3f019bb4137808f6e33fb7959be2f919d41e87b970399d57f9e65b9fe", "generation_seconds_batch": 21.495543014025316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7648122310638428, "target_z": 1.3281249283838352, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.6611274480819702, "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_high/5695aa3b8f891e87d79e60dee8274afef1fb0832a5eb238059601066db17e7c8.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "5702c066aab36926e0913f6490f24a7e86d16a2effc0e491e36b835177e6d18f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9b974b1926868927e419a1b770147940f196e3d9a02d3b27018328b5ac703814", "generation_seconds_batch": 17.9918425999349, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.805161237716675, "target_z": 2.578124928383835, "ce": 6.252951145172119, "genuineness": 3.8671910762786865, "identity": 0.2878921329975128, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5702c066aab36926e0913f6490f24a7e86d16a2effc0e491e36b835177e6d18f.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "570898f577b36e2bde7e5ebcd47fc8d10d376edbcad8d7d68459ec2f81a08d62", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6fc4f109c0bd06712275382e5dfdb04119533bdf15d0c835b20d9f57e828d1a5", "generation_seconds_batch": 22.494268746930175, "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": 6.221180438995361, "genuineness": 4.425758361816406, "identity": 0.563776433467865, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/570898f577b36e2bde7e5ebcd47fc8d10d376edbcad8d7d68459ec2f81a08d62.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "57150869018c04d243ec546eb920d9514284a51c5d696cf1d93c743e59e02b7f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4fd103d211c682174d07ea3bcb85b2c0f89e006b4c87de31b050b0974712c2dd", "generation_seconds_batch": 22.272687204065733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4410390853881836, "target_z": 2.140624928383835, "ce": 6.054384231567383, "genuineness": 4.068024158477783, "identity": 0.5075384378433228, "hypothesis": "Ik heb het kleinendoosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/57150869018c04d243ec546eb920d9514284a51c5d696cf1d93c743e59e02b7f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "57c1d766940290901e322a037365a0485053cfc4633f116b7e130d6e12a2059d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "948d0ee28426f577379e177342970c47b8f91e96df9218e476fc8e2105a99a33", "generation_seconds_batch": 16.62819466891233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.946873426437378, "target_z": 1.5468750716161648, "ce": 6.332377910614014, "genuineness": 4.419482707977295, "identity": 0.6133551001548767, "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_high/57c1d766940290901e322a037365a0485053cfc4633f116b7e130d6e12a2059d.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "57e39af03b10f9d91836510be06c6fcc57b03fc28ddea80eb0adacb9a2df6ba8", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e9c6fa74fb47da9a075cefe511ffd631bb4dd5d960f613ed7297ab6ace7011c1", "generation_seconds_batch": 17.971125644049607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7244632244110107, "target_z": 0.07812492838383514, "ce": 6.340320587158203, "genuineness": 4.570107460021973, "identity": 0.5488982200622559, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/57e39af03b10f9d91836510be06c6fcc57b03fc28ddea80eb0adacb9a2df6ba8.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "580e9a4ae8e35c55a31a9c66e049603495bff294ddee250bc948881f3affd978", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "99f20db4373995b14c6d5ccbf59f78cac82b91de1d2b4abab1d56fd96542abf9", "generation_seconds_batch": 17.257350855972618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5580782890319824, "target_z": 2.2812498567676704, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.5522418022155762, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/580e9a4ae8e35c55a31a9c66e049603495bff294ddee250bc948881f3affd978.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "582d9ee0e8bd24ebc9effcc0fe4a484f36d899e872064925f42a6b4b307050f2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "dfcea8d64ca49a2917179737ad58711f9b92688a388b6d0e94402c4365224c36", "generation_seconds_batch": 16.62819466891233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1159300804138184, "target_z": 1.75, "ce": 5.327630043029785, "genuineness": 3.9550557136535645, "identity": 0.4776327610015869, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/582d9ee0e8bd24ebc9effcc0fe4a484f36d899e872064925f42a6b4b307050f2.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "5880be67b712708f3f59ed8db3071c91df32d61b3b3d0f5b705c5b6b5198345c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fa9aad33f709cd485969f72df54241ad24a6616c6aa5094f158d52d1665dbdf2", "generation_seconds_batch": 29.81618393503595, "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": 6.221180438995361, "genuineness": 3.9676074981689453, "identity": 0.632104218006134, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5880be67b712708f3f59ed8db3071c91df32d61b3b3d0f5b705c5b6b5198345c.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "5880d3923ea67f15cd3cd15c96ee0652beac9ca1b27c6162e5c2a6ee36309eac", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "19ffcf804d3ea5089ad109655c2b43311a4e7e8ca0b91ee7e2cd2f505424c77d", "generation_seconds_batch": 15.188824643963017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7789191007614136, "target_z": 0.14355474568813395, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.6742358207702637, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5880d3923ea67f15cd3cd15c96ee0652beac9ca1b27c6162e5c2a6ee36309eac.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "58fd8fbdfc0a16b3b31ae55b39e5db8aaafb80ea9531b120888d0f9f3a18af8c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "981cffac7e2a709a2c0d3d3579dd70c9b1f8acd8291ea79d44f73ca746bf3aac", "generation_seconds_batch": 18.118048656033352, "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": 6.165581703186035, "genuineness": 4.268857479095459, "identity": 0.29673412442207336, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/58fd8fbdfc0a16b3b31ae55b39e5db8aaafb80ea9531b120888d0f9f3a18af8c.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "58fe0ced2411ed80fd4c31964a270b60d8ba6d57e2b5005b7c024c350b7e0560", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3632efbecde49f5e5d241b87f623d25c7d0c07948206ff9d5581e7b47210c77d", "generation_seconds_batch": 24.513826826005243, "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": 6.284721851348877, "genuineness": 4.582659721374512, "identity": 0.8328214287757874, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/58fe0ced2411ed80fd4c31964a270b60d8ba6d57e2b5005b7c024c350b7e0560.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "591604635125c5323a985692e24007aa20d0c8b6cd8b021bf8efdafe5e4b6009", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0f27e8e507a6c983164477e502854942c3970c0c2deb8b591ed475a0554770ab", "generation_seconds_batch": 22.484702782938257, "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": 5.927301406860352, "genuineness": 4.432034492492676, "identity": 0.8514101505279541, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Espéra-", "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_high/591604635125c5323a985692e24007aa20d0c8b6cd8b021bf8efdafe5e4b6009.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "5934535be095016d34720c4faf2827ed5247f94bfeab8b7e89488133e0e1e65e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ee3a07c7a21970a151a9c771565fac84a33e4b99346d5620dddc00ff6eaee094", "generation_seconds_batch": 25.0606854299549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9923887252807617, "target_z": 1.6015626074242473, "ce": 5.708878040313721, "genuineness": 4.356721878051758, "identity": 0.5621751546859741, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5934535be095016d34720c4faf2827ed5247f94bfeab8b7e89488133e0e1e65e.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "595f3f985bb694162b73064e098f271c45c505bf8845b28c0813c16672115bfd", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1a2a7fc7ee939d75ab367080bf190c1388d447227bbc09fb661feaa5d06640e1", "generation_seconds_batch": 18.092408030992374, "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": 6.1894097328186035, "genuineness": 4.6203155517578125, "identity": 0.6293108463287354, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/595f3f985bb694162b73064e098f271c45c505bf8845b28c0813c16672115bfd.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "59a1b7cc879dde88d280a5e3bb178df6085306a1965446ce6c59da7ebc7d2925", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fadf4d999876912ab7f2c67a955a897e3414d832108fe326e864776b83cd1442", "generation_seconds_batch": 17.628292661043815, "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": 6.372091293334961, "genuineness": 4.331617832183838, "identity": 0.4027724862098694, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/59a1b7cc879dde88d280a5e3bb178df6085306a1965446ce6c59da7ebc7d2925.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "59b3aabc0b1c07a0fd5fc1338ba507c6a73d68ae2a01fbf263ca7939ac4d2805", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f1268959165149991decbecb0d699c998b50890d383fe0a82ea23abbd365b38b", "generation_seconds_batch": 22.922349330969155, "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": 6.252951145172119, "genuineness": 4.595211505889893, "identity": 0.5801441073417664, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/59b3aabc0b1c07a0fd5fc1338ba507c6a73d68ae2a01fbf263ca7939ac4d2805.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "59c2127aa2a9cfe9b4e44bf261f7f7731436d1a2973265176c8835375646071e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "daf5ed1f72c8cf676a28e563eae2ccefa6ec1317e5fa630151015fdd96b6f907", "generation_seconds_batch": 22.538021008018404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7713146209716797, "target_z": 1.335937679040412, "ce": 5.510311126708984, "genuineness": 4.281409740447998, "identity": 0.4380602240562439, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/59c2127aa2a9cfe9b4e44bf261f7f7731436d1a2973265176c8835375646071e.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "59f531c9db2c74f3dca5a34dbd099b4bb194b9b5bd4878cb2d5e0d82faca25d0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8fa099fea4073b42be0ec22ea4afd6b5d79cb29324838790d9818053ec366887", "generation_seconds_batch": 23.20532251300756, "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": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.764598548412323, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/59f531c9db2c74f3dca5a34dbd099b4bb194b9b5bd4878cb2d5e0d82faca25d0.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "5a3977583cc37a4b44bb4a494af20b2a3ce44e941b89e11c62b106b3ddb36145", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c6ddaad2266c92ca2766589413260317b879e32263bfb207ffb9e01747fa409b", "generation_seconds_batch": 21.577313233050518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.790821075439453, "target_z": 1.3593750716161648, "ce": 6.0146708488464355, "genuineness": 4.293961524963379, "identity": 0.5132668018341064, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5a3977583cc37a4b44bb4a494af20b2a3ce44e941b89e11c62b106b3ddb36145.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": "5a521e13ef7c1caa8d005a13372e53d9d0d47c235598bbefee29605683070571", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11dc65459ba31eba677e098e8a78119850f65986e672e8934a1b22186892b21d", "generation_seconds_batch": 17.465600084979087, "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": 6.1894097328186035, "genuineness": 4.093128204345703, "identity": 0.5606008172035217, "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_high/5a521e13ef7c1caa8d005a13372e53d9d0d47c235598bbefee29605683070571.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "5a89691d99f6478d6e67692798c1a4b242421f3807ff2ec4807bb5771767f5d3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9b67797932db366178fbd647b97c4d5141d7822e795951769d895d9a2106c705", "generation_seconds_batch": 24.371148369973525, "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": 6.316492557525635, "genuineness": 4.68307638168335, "identity": 0.7147443294525146, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5a89691d99f6478d6e67692798c1a4b242421f3807ff2ec4807bb5771767f5d3.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "5ab7e620ace8270f2be21a33268d1a7527982b23cea5abc2d9d8e6fc032e1a90", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bfa1c2c97c6aaec1e18d92274be8dfd3dc7b323dc980d6faab881444ba9a4472", "generation_seconds_batch": 17.75679139792919, "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": 6.133810997009277, "genuineness": 4.218649387359619, "identity": 0.47863298654556274, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5ab7e620ace8270f2be21a33268d1a7527982b23cea5abc2d9d8e6fc032e1a90.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "5ae12633ec2099b1f47f048f2afefd149f4e4832b91e1169c01195e66dbae5b2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "87bbbeb7aea10a4e5503518281f58a62e74a8fdc21fabd37552c8e8a6ef7774c", "generation_seconds_batch": 18.095353911980055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065248489379883, "target_z": 2.890624928383835, "ce": 6.221180438995361, "genuineness": 3.7165660858154297, "identity": 0.4492162764072418, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5ae12633ec2099b1f47f048f2afefd149f4e4832b91e1169c01195e66dbae5b2.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "5b1c5e70424e220f9123c976f1b5eaf1c5ad85c51622b72718b1b21c8c8cc3a8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9fc3a675a8462546d7a5f713d1e50b9512879857f004273485505b84ec038aa2", "generation_seconds_batch": 29.81618393503595, "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": 6.1894097328186035, "genuineness": 3.9927117824554443, "identity": 0.2675977945327759, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5b1c5e70424e220f9123c976f1b5eaf1c5ad85c51622b72718b1b21c8c8cc3a8.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "5b64ed3c33f21f6ac702be68cd27d7cda13e5e9706d1d40c2e632e2ccb185bfd", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "88fe686c8d3a2f891afb89ab9696785c954465c361a2049407cbec93c36b69c8", "generation_seconds_batch": 17.206087893922813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4627504348754883, "target_z": -0.23632818766414426, "ce": 6.372091293334961, "genuineness": 4.570107460021973, "identity": 0.39154380559921265, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5b64ed3c33f21f6ac702be68cd27d7cda13e5e9706d1d40c2e632e2ccb185bfd.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "5b82c5ccea441f3e5166e1899232121461b97e16ccbf24553c2508fd58942b1c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f354c6c687e84c9d2a951fee494542ce4d68c603f53e6363db8bc8f260120585", "generation_seconds_batch": 18.59255115792621, "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": 6.419747352600098, "genuineness": 4.519899368286133, "identity": 0.4246334731578827, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5b82c5ccea441f3e5166e1899232121461b97e16ccbf24553c2508fd58942b1c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "5ba219e168e800c6167356b522022d61915af42620b20e3caf7b947a12ecd34e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "47fa2a90b398ba854747142988de3497a995ed35d61ed7036bff721f1f16e2fd", "generation_seconds_batch": 16.32634205603972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.750472068786621, "target_z": 0.10937507161616486, "ce": 6.1020402908325195, "genuineness": 4.356721878051758, "identity": 0.7090169787406921, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5ba219e168e800c6167356b522022d61915af42620b20e3caf7b947a12ecd34e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "5baf3f21d5f537a6b91af7af210833ad70cb98276f73fe612ea043e2e9076a62", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d4ad1df3c4bccbed255d1484c80b376f2c26380457509e791603f365e6142e33", "generation_seconds_batch": 16.242419044021517, "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": 6.427689552307129, "genuineness": 4.532451152801514, "identity": 0.5281739234924316, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5baf3f21d5f537a6b91af7af210833ad70cb98276f73fe612ea043e2e9076a62.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "5be0139e30904e6027d7f3ee21e159743df90c4bff877775b0528b3e3dbf26fb", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "39169267e2e66d3bf981a97d799d493e61e2adf44d6fa6d82784a9d84f597cba", "generation_seconds_batch": 17.055181364994496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.576249122619629, "target_z": 1.1015626074242473, "ce": 6.046441555023193, "genuineness": 4.344170093536377, "identity": 0.3992936611175537, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5be0139e30904e6027d7f3ee21e159743df90c4bff877775b0528b3e3dbf26fb.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": "5c4014c2b23444b4b8e71c8845ca7a5d2114b865356c7ff98d99b36ab8d37b6d", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec72ee9c22ed891ae685c3649da964b9da0a8930e68bd6850b4114e64610801d", "generation_seconds_batch": 23.88780442602001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.831170082092285, "target_z": 2.609375071616165, "ce": 5.998785495758057, "genuineness": 4.143336772918701, "identity": 0.5843614339828491, "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_high/5c4014c2b23444b4b8e71c8845ca7a5d2114b865356c7ff98d99b36ab8d37b6d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "5c6eaa69d4b13120ac039abd65988ff21dabb95631e0ed544d27130cf3c212a5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f0f58f8f4db08163e5fcb4bd4e6c8a2a30b9ee886300a74c30810bda811875f4", "generation_seconds_batch": 18.095353911980055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.663449287414551, "target_z": 3.609375071616165, "ce": 6.181467056274414, "genuineness": 3.8169827461242676, "identity": 0.4757576584815979, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5c6eaa69d4b13120ac039abd65988ff21dabb95631e0ed544d27130cf3c212a5.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "5cac8ed0e93177f30eb29a3861211016fb17bada402dc4459b114d51ffffc218", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4860a63015a0b883937925b4d85ae450b345bb99098fef8d4fd54958d31397d0", "generation_seconds_batch": 18.31541846797336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.724869728088379, "target_z": 0.0786133506281597, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.7363302707672119, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5cac8ed0e93177f30eb29a3861211016fb17bada402dc4459b114d51ffffc218.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "5cff9167963e9d8e299cd207535c634cc25034aaa2ae6adf0272b17c6bbede1c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "fa5374d14f34ee21b989134b8eac44da1a370633007903426c904461bd549c20", "generation_seconds_batch": 23.144915941986255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6139261722564697, "target_z": -0.05468753580808243, "ce": 6.1020402908325195, "genuineness": 4.482243061065674, "identity": 0.542371392250061, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5cff9167963e9d8e299cd207535c634cc25034aaa2ae6adf0272b17c6bbede1c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "5dad0a53b83cda6ab03c95a1b767ce8af492e6b24bba1c10888c5515a66c74f8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "419b4eb7be3b00353b019d52b75aa2432faee81a157fed398ceb46693cdd73ae", "generation_seconds_batch": 23.4522893599933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.208296775817871, "target_z": 3.0625002864646595, "ce": 4.984109401702881, "genuineness": 3.747946262359619, "identity": 0.669737696647644, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5dad0a53b83cda6ab03c95a1b767ce8af492e6b24bba1c10888c5515a66c74f8.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "5de9e1c32cc3616a1e81c03a7131567f0b07f68ae581fbe251ba6ba10b638a16", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0bb013196bc95a6ff35c705d0143231e3c01c0cb0cf765a4a0e6d224831ab30f", "generation_seconds_batch": 22.969574319082312, "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": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.4304012060165405, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5de9e1c32cc3616a1e81c03a7131567f0b07f68ae581fbe251ba6ba10b638a16.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "5dfffcb203559046c72cfe5bea3613a037ab3203eab798eff8738b3a1128526d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "feca33308943b2e52e05895d36ddff8efc96dfabbb277f7ab6a8f54472a3548d", "generation_seconds_batch": 22.20982741599437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6724457740783691, "target_z": 0.015624928383835142, "ce": 6.022613525390625, "genuineness": 4.325342178344727, "identity": 0.5164715051651001, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5dfffcb203559046c72cfe5bea3613a037ab3203eab798eff8738b3a1128526d.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "8c2ea711ef15d53fcd3383cd6b2ffd075b8db7726e1de0f6a3e542702214b690", "generation_seconds_batch": 17.84025018394459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868847131729126, "target_z": 1.4531249283838352, "ce": 6.0305562019348145, "genuineness": 4.118232727050781, "identity": 0.7012172937393188, "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_high/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "5e5680914d1a47767ddbf4012a1ff8a18fa05a5143a9d84668cf0a7a62538317", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "370ba7247da6e1d7c836d0317f495d66ee6fec7b6b6facae63f5d98bbf089bce", "generation_seconds_batch": 23.933879681979306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940371036529541, "target_z": 1.539062320959588, "ce": 5.7684478759765625, "genuineness": 4.545003414154053, "identity": 0.4816887080669403, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5e5680914d1a47767ddbf4012a1ff8a18fa05a5143a9d84668cf0a7a62538317.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "5e93ca6f48c4ae8ed8e426a79998962dc9c9ec7477249202680033e4add1e99d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9b73d2d334c116605e20aac5004c27b9154cc74a30d577956ba51058377e1ae2", "generation_seconds_batch": 21.800078919972293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823331832885742, "target_z": 1.3984373925757527, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.38788822293281555, "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_high/5e93ca6f48c4ae8ed8e426a79998962dc9c9ec7477249202680033e4add1e99d.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "5f51d314865a35e7c1161858f171fd854f6a91517fba55178efe9f3e7f69baf2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bcbb9f852da8e83d902b6bc441dce6b724b906311cbe280f4f35ce065d4125e7", "generation_seconds_batch": 23.933879681979306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2608933448791504, "target_z": 0.7226563753282885, "ce": 5.186647415161133, "genuineness": 4.306513786315918, "identity": 0.3819623589515686, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "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_high/5f51d314865a35e7c1161858f171fd854f6a91517fba55178efe9f3e7f69baf2.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "5f5ae0aefbc9516a9d25a091e557c3aa16b6f191fdf2261640f22fd1f6caed04", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "991e23be9f656c01e8ff666c5adaf1cd4533783926c200902b285a8367eb3878", "generation_seconds_batch": 15.984864283003844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6639117002487183, "target_z": 0.00537106913194333, "ce": 6.157639026641846, "genuineness": 4.7332844734191895, "identity": 0.7921465635299683, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5f5ae0aefbc9516a9d25a091e557c3aa16b6f191fdf2261640f22fd1f6caed04.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "5f64ff77f72f4bac1d5b0dfdbeb6749aa42d6c1fcf09cf954a9fb0582ec2ad4d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "de66b282267e3bd106b7bc00aef6a9210f540dca486c2e38a44ea578cbbb4073", "generation_seconds_batch": 21.999378002947196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.624436378479004, "target_z": 3.5625002864646595, "ce": 5.919358730316162, "genuineness": 3.9927117824554443, "identity": 0.559878408908844, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5f64ff77f72f4bac1d5b0dfdbeb6749aa42d6c1fcf09cf954a9fb0582ec2ad4d.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "5f740645f5e8fc68b0c482f71c9d07d6dc6738b6eca24d31caba784fa80c7855", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3fa443e6109af7ad1415585a86f8664fe5b409ccb0d9621fa693454b979466ed", "generation_seconds_batch": 22.256560069974512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2478888034820557, "target_z": 0.707031160479794, "ce": 6.133810997009277, "genuineness": 4.507347106933594, "identity": 0.30268165469169617, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5f740645f5e8fc68b0c482f71c9d07d6dc6738b6eca24d31caba784fa80c7855.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "5f98f13bed42bce9b7597ee642169d8d3c1a7f41882dd0e772fd5e930aebe3a3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cf46bd081dd3b07fbf12ae840b8e09a4b3bd9ecbf7e72a1269e282948062ffe5", "generation_seconds_batch": 24.48482840601355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.979384422302246, "target_z": 1.585937679040412, "ce": 6.078212261199951, "genuineness": 4.293961524963379, "identity": 0.6988277435302734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5f98f13bed42bce9b7597ee642169d8d3c1a7f41882dd0e772fd5e930aebe3a3.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "5fd5b82f0e17a79fe40d3e23a606be251e8cf865295e8b003ee854a08e9adbc4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f097b3cc02fd02825400f7238bbf6fc6353e8965e348cb305518c39a82edac2b", "generation_seconds_batch": 25.212228736956604, "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": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.5428310036659241, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/5fd5b82f0e17a79fe40d3e23a606be251e8cf865295e8b003ee854a08e9adbc4.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "61106874c43cde2905425a1c7be06c310f6155d1d348024392f1c18ddcdb7881", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "19368f6a908bee3782078681bd109b4c510db9dde578c90ab10889648dcd5f9a", "generation_seconds_batch": 17.389430526993237, "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": 6.229123115539551, "genuineness": 4.7081804275512695, "identity": 0.4452303349971771, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/61106874c43cde2905425a1c7be06c310f6155d1d348024392f1c18ddcdb7881.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "6112901e0713d55de69feb647319c1075d7a3cb3a3884b47789db7454241a28f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1ea7255a770f63324b7e40e0ca307901f55ee2f002f872e567fadcfac32aeade", "generation_seconds_batch": 23.921840654104017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.088585376739502, "target_z": 0.5156249283838351, "ce": 6.252951145172119, "genuineness": 4.507347106933594, "identity": 0.5714362263679504, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6112901e0713d55de69feb647319c1075d7a3cb3a3884b47789db7454241a28f.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": "612247258d9a13ccd5e2768f80af53788ba676488dd6e35a93b390f5666c2232", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89ebc78fd395c5063df258ddbce1f4c1742d7b7937055881f9fa349d7f6a1840", "generation_seconds_batch": 23.107930066995323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4410390853881836, "target_z": 2.140624928383835, "ce": 6.308549880981445, "genuineness": 4.2249250411987305, "identity": 0.40259045362472534, "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_high/612247258d9a13ccd5e2768f80af53788ba676488dd6e35a93b390f5666c2232.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "614b84bb844dfe0bfc70a6c98ea51c99ee932a9a0f7181362d9e66b52324a946", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ad26ffcd5cb94bf784e0f671eee705a0f9ed4706de54cf24a14b2a5dbbe0a111", "generation_seconds_batch": 18.26535903196782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.381183624267578, "target_z": 0.8671875358080824, "ce": 6.316492557525635, "genuineness": 4.7458367347717285, "identity": 0.7096911072731018, "hypothesis": "He encontrado la. He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/614b84bb844dfe0bfc70a6c98ea51c99ee932a9a0f7181362d9e66b52324a946.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "6154531ba56525ae09329e67fe6fed6fe101747809f421b2b0f533c1d9c66bed", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1df92b75300c3f3fe2c78152df0c35e9cf10577840fd5dce46fd700506d74289", "generation_seconds_batch": 16.797234767931513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.703940749168396, "target_z": 0.053466766661930454, "ce": 6.348263263702393, "genuineness": 4.570107460021973, "identity": 0.35319089889526367, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6154531ba56525ae09329e67fe6fed6fe101747809f421b2b0f533c1d9c66bed.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "61ad1cb41435d48dc41cf6a6557ddea509a264ce0ec4c32d9553e85dec3db18d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c9c8dc59d0505c3ac09032be29dc28f15152598ce039a090b1b9889aeaa6427f", "generation_seconds_batch": 22.93818044802174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4410390853881836, "target_z": 2.140624928383835, "ce": 5.720792293548584, "genuineness": 4.406930446624756, "identity": 0.6347808837890625, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/61ad1cb41435d48dc41cf6a6557ddea509a264ce0ec4c32d9553e85dec3db18d.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "61d4a70596285d6888ac6b1297e5a5a86ba4dafb3c759a6cfff196c68e25934c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "103d51b866ad30d7568fd480f29369fadfe1513ec5466702e57f9528875c3e77", "generation_seconds_batch": 21.823137464001775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.819501876831055, "target_z": 3.796875358080824, "ce": 5.49045467376709, "genuineness": 3.6977379322052, "identity": 0.6241198778152466, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/61d4a70596285d6888ac6b1297e5a5a86ba4dafb3c759a6cfff196c68e25934c.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "61e6de40b359aea9b35b34e3a3223d33e000820f823d3f37806b40d7f6aee256", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "405c8bf5cbf890be57ce54f0e0d1aed9731351b240c6f12e30b0209e1d951443", "generation_seconds_batch": 22.48021783796139, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2264671325683594, "target_z": 1.8828124641919175, "ce": 5.1052350997924805, "genuineness": 3.9425034523010254, "identity": 0.5342209339141846, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/61e6de40b359aea9b35b34e3a3223d33e000820f823d3f37806b40d7f6aee256.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "6286558bee97176732406e4c088a0d3b43afba3ccc4b5551535423cd9e9b1b6d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ddffeb2397de34d94afab5cf040c5c3a2299145105a61f6fda41ceda62eee339", "generation_seconds_batch": 22.56464088591747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.571082830429077, "target_z": 2.296875071616165, "ce": 6.165581703186035, "genuineness": 4.080576419830322, "identity": 0.6180362701416016, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6286558bee97176732406e4c088a0d3b43afba3ccc4b5551535423cd9e9b1b6d.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": "62a399c6c5bdc3353aefed5b8e7b9228be41ad7f3384f765815f250e590d5918", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07c875a247446d039c6e8db5b019f382f2982ec51f0cf5057be24ed0765e9451", "generation_seconds_batch": 21.365181021974422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.727135181427002, "target_z": 2.484375071616165, "ce": 6.062326908111572, "genuineness": 3.9425034523010254, "identity": 0.27285152673721313, "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_high/62a399c6c5bdc3353aefed5b8e7b9228be41ad7f3384f765815f250e590d5918.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": "62cc9bfe2ecdcc759e66fcef9d230585af81ce2eb06b3a5897699d1e21c559fe", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8512aeb416623466b04d23cdf203b94e0921ef7bf19ecf0fdacfcb38879405cf", "generation_seconds_batch": 22.89692713401746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6542751789093018, "target_z": 1.1953124641919175, "ce": 5.935244083404541, "genuineness": 4.545003414154053, "identity": 0.6296374797821045, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade. Tann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/62cc9bfe2ecdcc759e66fcef9d230585af81ce2eb06b3a5897699d1e21c559fe.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "62ce8356c24dfe762219a3910d08fe5cb9c1f5ff40bf80e75331cf592f950747", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cd1b1030fb8ec41c3791fbbffb03d5772792ddad09d866f7fcc1160eaaff55d2", "generation_seconds_batch": 22.10518890491221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2524757385253906, "target_z": 1.914062320959588, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.4473586678504944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/62ce8356c24dfe762219a3910d08fe5cb9c1f5ff40bf80e75331cf592f950747.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "6326f24a88f039873e2f060c2e419e0bb9a7d70745051eb9bfbac9e225147a53", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "68261ecb9cfa19c06886a317e67f806b52974a55cc591e864a0a9119e04a152b", "generation_seconds_batch": 23.904840764938854, "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": 5.788304805755615, "genuineness": 4.218649387359619, "identity": 0.5612652897834778, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6326f24a88f039873e2f060c2e419e0bb9a7d70745051eb9bfbac9e225147a53.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "63758597e8b35172f450a70b900ebcedb5f3982df32ef20e7a0a493c99ab4f42", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "45684b5efcf4fb67ca8045301b8185faea022bf01b63d14532e047b9b65b1af4", "generation_seconds_batch": 22.488092345069163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467047929763794, "target_z": 2.171875071616165, "ce": 5.824047088623047, "genuineness": 3.9801597595214844, "identity": 0.34750938415527344, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are right.", "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_high/63758597e8b35172f450a70b900ebcedb5f3982df32ef20e7a0a493c99ab4f42.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "63f45f16ba93acf6a110bb750592ad00522b589c6f1b76a0255cc325e58af7b4", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "14b9bb64f944432933fac08f291b34e0069cd4e41fcd53eeba48a42f3838d551", "generation_seconds_batch": 17.573196347919293, "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": 6.09409761428833, "genuineness": 4.268857479095459, "identity": 0.5073193907737732, "hypothesis": "Ik heb een klein doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/63f45f16ba93acf6a110bb750592ad00522b589c6f1b76a0255cc325e58af7b4.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "64a05b364b8bb28f5a72535c7f097c25252b61265a637d812ace1c2b9f30967f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a25a211847768010395bf678c04bdb70fa7f7ef04772c76e346e244be5bb10d8", "generation_seconds_batch": 16.93509841291234, "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": 5.585766792297363, "genuineness": 4.432034492492676, "identity": 0.7530373334884644, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/64a05b364b8bb28f5a72535c7f097c25252b61265a637d812ace1c2b9f30967f.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "64b5b4c546fb0d3e042c4b1975433d82b9f32431608cf6ea624faba027ac21fb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "264bdf1cc657572c0b2d58cff0a1aa228f4e4d540729af008695147bdfa6263c", "generation_seconds_batch": 18.799026050022803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.303157329559326, "target_z": 0.7734373925757527, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.8050892353057861, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/64b5b4c546fb0d3e042c4b1975433d82b9f32431608cf6ea624faba027ac21fb.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "64c426171e1b88c8c854ad67b9b396863db1417289aac669ac393ea9abcadbbb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b38ec47f10b262e40e865debcae14ee96e5f65cc822e88fa791a1ddf719056b2", "generation_seconds_batch": 17.201562503003515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8558428287506104, "target_z": 1.4375, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.6430745124816895, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/64c426171e1b88c8c854ad67b9b396863db1417289aac669ac393ea9abcadbbb.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "651e19a41d1b79876efdbb99c640fed0ea78b78d4e9913eb2fbbf59db7658cf3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5892bcd2008870a050dd038bedafa4732561ad2dd5f200e72fb994058f55fa1f", "generation_seconds_batch": 17.964187899022363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.814681053161621, "target_z": 0.1865234419760103, "ce": 6.117925643920898, "genuineness": 4.7081804275512695, "identity": 0.5938077569007874, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/651e19a41d1b79876efdbb99c640fed0ea78b78d4e9913eb2fbbf59db7658cf3.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": "65957ed86b44da91818d17034e4db5dd9acd49ff1f5c8f8174ccc09eb1efff97", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b9d74e7a1a5d0538102a4e0fb2892180b63e985e27e9e3643fe61a323eb5c6e", "generation_seconds_batch": 21.365181021974422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.797323226928711, "target_z": 1.3671875358080825, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.6022577285766602, "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_high/65957ed86b44da91818d17034e4db5dd9acd49ff1f5c8f8174ccc09eb1efff97.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "65f2caec80d2c3c2a4b03524c39d8a33911eb859878c9973c98352de92cf107a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f480f67f094b61fb01b641310bc7da487b188ad504b5238f3d9290afeb22d568", "generation_seconds_batch": 22.67225395492278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.546410083770752, "target_z": 3.4687501432323296, "ce": 6.109982967376709, "genuineness": 3.7793264389038086, "identity": 0.6293301582336426, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/65f2caec80d2c3c2a4b03524c39d8a33911eb859878c9973c98352de92cf107a.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "6668007451f01508b1cb88e2d7b60d7b665321346d392c6fc801c4d595bfab26", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2486dcc74a2ddd60ddd28c9ea3a353d9f94cbb7f9c775791ca37a8409eca0509", "generation_seconds_batch": 17.35070348205045, "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": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.6624724864959717, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6668007451f01508b1cb88e2d7b60d7b665321346d392c6fc801c4d595bfab26.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "66d2e0746775293fd835273a39720df847747ff0f08ffc7dae006d8ca22f687f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2144422402efc3d230ae98fdce66f9f8b4a43577ba20fc2142e4bffe2a5cec90", "generation_seconds_batch": 23.109901037998497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2576422691345215, "target_z": 0.7187501432323297, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.39764344692230225, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/66d2e0746775293fd835273a39720df847747ff0f08ffc7dae006d8ca22f687f.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "672df1d7525f3c52ec5e2ba4d6a89b877763d2d0530c9970343c4a62e5b2f270", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ba85467d45332a7ed4c31899e8fa500015ff932977cdd5123109779df332da37", "generation_seconds_batch": 18.1167323050322, "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": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.6146609783172607, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/672df1d7525f3c52ec5e2ba4d6a89b877763d2d0530c9970343c4a62e5b2f270.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "67bf58ef25fa6f537170b3b3890b4fd41e07a78a6d9f4ad443cbf1879ce4b7a1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "97840660ce8bb77cefcb6d44539a85767fe8b83a3773bdd869a84f0f3c96ba0c", "generation_seconds_batch": 21.508463278063573, "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": 6.221180438995361, "genuineness": 4.017816066741943, "identity": 0.27957767248153687, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/67bf58ef25fa6f537170b3b3890b4fd41e07a78a6d9f4ad443cbf1879ce4b7a1.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "67d5bae40a7eab4dbdbfd54de4c76f2f6029cdabc1a4b3a9f52dff61d46bce75", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e9b6f26a4cf2067d346ecbbbba0c7175be0ed868222fe65f65a7cb5e69f5dce4", "generation_seconds_batch": 23.341008462943137, "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": 6.197352409362793, "genuineness": 4.7834930419921875, "identity": 0.430223286151886, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/67d5bae40a7eab4dbdbfd54de4c76f2f6029cdabc1a4b3a9f52dff61d46bce75.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "67eb5bb3f4702c068dc369fb6e5251183b05e0019e0011a79e29763f1b31fab2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c0631d1b0161701f7f448d38e9f5b8b5588bb71c046c16635d5fbe963ae29ca7", "generation_seconds_batch": 23.825047438964248, "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": 6.308549880981445, "genuineness": 4.444586753845215, "identity": 0.5518380403518677, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/67eb5bb3f4702c068dc369fb6e5251183b05e0019e0011a79e29763f1b31fab2.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "681ffb7966233a72a287e21a60cd9ad0400cdfec1e4c3c0ed3c8caca2292588a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7c18d7daadf1cb39a5b85a9ae02001cde7bb6a4ba8de4c5bcc526b90b742849b", "generation_seconds_batch": 22.676186230033636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9988908767700195, "target_z": 1.6093750716161648, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.7929439544677734, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/681ffb7966233a72a287e21a60cd9ad0400cdfec1e4c3c0ed3c8caca2292588a.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "68320d10af66eafd0b8f2c679d25e8e66b6c310e821edadaeb834cb2c57f97cb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5f1caf5ba3138c203ae721937d28d870cbcb2ab0cd815a8bd2058c74cd1b05e6", "generation_seconds_batch": 17.306138889980502, "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": 5.927301406860352, "genuineness": 4.607763767242432, "identity": 0.834865927696228, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/68320d10af66eafd0b8f2c679d25e8e66b6c310e821edadaeb834cb2c57f97cb.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "68a36fc9e47b48ad411ab8c0605219614591ddaa8b4e3958f65575ceff8f1291", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f674e75edac0d0fe64a4a67ebebae4b9e459d44139d80ce2150c9f30375a1a5", "generation_seconds_batch": 17.290536138927564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.202373504638672, "target_z": 0.6523436246717115, "ce": 6.24500846862793, "genuineness": 4.18099308013916, "identity": 0.3531162142753601, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/68a36fc9e47b48ad411ab8c0605219614591ddaa8b4e3958f65575ceff8f1291.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "68c0fa911fa77fc6567cef8baf74fff85e429f8be503ca72a4a8a05ce61c647a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a6cfb48efbefed41bb62ec974ae7816dd9b8fa6b29c9f844504e5734c1cbc5aa", "generation_seconds_batch": 16.67468790698331, "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": 6.316492557525635, "genuineness": 4.6956281661987305, "identity": 0.6050167083740234, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/68c0fa911fa77fc6567cef8baf74fff85e429f8be503ca72a4a8a05ce61c647a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "68e0887328cef3aae9a1d2dd35a930af1c647f23bd2115fb5779865a5a41ac00", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "33198190de0122a1b65c31c51b18933154a1e6c07307e229f4a7023f9645260f", "generation_seconds_batch": 23.921840654104017, "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": 6.205295085906982, "genuineness": 4.7081804275512695, "identity": 0.5302175283432007, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/68e0887328cef3aae9a1d2dd35a930af1c647f23bd2115fb5779865a5a41ac00.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "6909b0889fdb144c64a846cf4e51e987400adf0a2abc2202be1c8f57d5a8de0f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fde2a2939832322c165ab3def0f9c0ad63ffde9e04519bbc3068e7210832211a", "generation_seconds_batch": 24.43296765605919, "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": 6.260893821716309, "genuineness": 4.595211505889893, "identity": 0.5024849772453308, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6909b0889fdb144c64a846cf4e51e987400adf0a2abc2202be1c8f57d5a8de0f.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "6946baaee76fd97052459941a4ef85b9fa39c52ae40abdc75dd3b9f04cbb14cb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e3b6707ee5dc3f153e7f1f16990c4e46630fe31041dd780cfa0bd4cf6c59ab42", "generation_seconds_batch": 17.108321174047887, "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": 6.046441555023193, "genuineness": 4.7081804275512695, "identity": 0.5224393606185913, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6946baaee76fd97052459941a4ef85b9fa39c52ae40abdc75dd3b9f04cbb14cb.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "6999770f17a1ff1ad564dc90aec9a22059b948cd744703a13a449070e93f531d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9da5f2c6f5864c49986f3592b2008bf63f41aaeb27f4fe4801db5209a9bd8a8f", "generation_seconds_batch": 16.59198665106669, "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": 6.23706579208374, "genuineness": 4.6956281661987305, "identity": 0.6925894021987915, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6999770f17a1ff1ad564dc90aec9a22059b948cd744703a13a449070e93f531d.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "69cd43fe33b98999bd7fb209f80697a8c2dec5ef2bc41530f4f890e18dd37b78", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "57ed14393da87b79159dbff86c0c83aa5e23cf9136fb45e812089d831f75bb67", "generation_seconds_batch": 22.356071872985922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5502402782440186, "target_z": 1.0703124641919175, "ce": 5.395142555236816, "genuineness": 4.168440818786621, "identity": 0.5181666016578674, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/69cd43fe33b98999bd7fb209f80697a8c2dec5ef2bc41530f4f890e18dd37b78.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "69ef29fa584e0e0c3fab928b564fc6ee3b2ab1113cbe0687a647002edc053c27", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f4271b72c19d1214e5b22d3ec130f3fc9c7fc6b713e8843bf2bbd923d9a4ac0", "generation_seconds_batch": 21.445947820087895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9585418701171875, "target_z": 0.35937507161616483, "ce": 6.260893821716309, "genuineness": 4.507347106933594, "identity": 0.3562711179256439, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/69ef29fa584e0e0c3fab928b564fc6ee3b2ab1113cbe0687a647002edc053c27.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "6a1998d8991b44a09f7f25f467e1fa1574aaa1cc3e3051cea8874d6d76349cb6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "188343919289c8490ef841e989994f5d6335bb0f35d5a705fb4b02bff0869a00", "generation_seconds_batch": 17.975838333950378, "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": 6.117925643920898, "genuineness": 4.632867813110352, "identity": 0.4618777334690094, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6a1998d8991b44a09f7f25f467e1fa1574aaa1cc3e3051cea8874d6d76349cb6.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1f18663041adb5fe4984d21dec641b7e6f137f45eb99df0845362793339e1396", "generation_seconds_batch": 17.611844150000252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524231433868408, "target_z": 1.039062320959588, "ce": 6.316492557525635, "genuineness": 3.8797430992126465, "identity": 0.43604776263237, "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_high/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "6b1e066eff952802872d9c2301cead952720852d3dfab86c37439b7d013a4d71", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "56aec77451b5c19dda143e95d5876a52aea763160cb7b108d68d150b591e1e28", "generation_seconds_batch": 17.206736281048506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.901647686958313, "target_z": 0.291015580239897, "ce": 6.24500846862793, "genuineness": 4.821148872375488, "identity": 0.6911578178405762, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6b1e066eff952802872d9c2301cead952720852d3dfab86c37439b7d013a4d71.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1fbc203afce43be14170437f2d15f6e3abf95a563e74f2f5b73dc1e1da418fb3", "generation_seconds_batch": 16.865519181010313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.5131939649581909, "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_high/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "6b9d63258e6462bc5cf8334a8297dab3e710eb4f23f19450421fcc68f02b0112", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "42a5c8e634554c1c7549c3ac00eb54dfcd61a8c2c543337e3126909473e15b6b", "generation_seconds_batch": 17.151638868963346, "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": 6.292664527893066, "genuineness": 4.557555198669434, "identity": 0.7067315578460693, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6b9d63258e6462bc5cf8334a8297dab3e710eb4f23f19450421fcc68f02b0112.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "6beaeccc71ee4be29cf827bef551be3b708e1135b7522781ac721841b7dcd886", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cf5ea5e57e80b842ac75ce37e20e0e4dfa2640c103e87767be76215cbcde4516", "generation_seconds_batch": 22.608567847986706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.267395496368408, "target_z": 0.730468839520206, "ce": 6.022613525390625, "genuineness": 4.7332844734191895, "identity": 0.6833025217056274, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6beaeccc71ee4be29cf827bef551be3b708e1135b7522781ac721841b7dcd886.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "6bf30f812f19786e9bc13eee1e53ec6ee82e5bdb55ba74fc8157646643f42a8e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 129, "seconds_generated": 10.32, "mp3_sha256": "8c7d50bf50f78884fa10670380e957e95a0090c919640ecae503a490700ea77c", "generation_seconds_batch": 23.260423870990053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.423447847366333, "target_z": 0.917968839520206, "ce": 5.903473377227783, "genuineness": 4.469690799713135, "identity": 0.5538164973258972, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "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_high/6bf30f812f19786e9bc13eee1e53ec6ee82e5bdb55ba74fc8157646643f42a8e.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "6c18fea7601f9266c1a1dd4906eacbc72cb17c94702a0d3aa584f5f97546ba17", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "eb3bc8e4c14143dabcc1b7bc794fd382cadf29d6dfda84cb36df00a18c68ee43", "generation_seconds_batch": 24.025337156956084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.86751127243042, "target_z": 0.25, "ce": 6.435632705688477, "genuineness": 4.394378662109375, "identity": 0.47235774993896484, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6c18fea7601f9266c1a1dd4906eacbc72cb17c94702a0d3aa584f5f97546ba17.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": "6c355df3b1e8d0b109ee4434f1ad5347cb81353c05e3952d705aa6764debee19", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a60a6001ea63433406c6138eb4e09d7ae2daf9295c3d04cfa8089d81c006cbe2", "generation_seconds_batch": 17.969894291018136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.790821075439453, "target_z": 1.3593750716161648, "ce": 6.3641486167907715, "genuineness": 4.419482707977295, "identity": 0.49594569206237793, "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": 0.120000000000001, "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_high/6c355df3b1e8d0b109ee4434f1ad5347cb81353c05e3952d705aa6764debee19.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "6d3cd42552fa31a88aa5a973cdd67c88b067509eac310957a7b6267aa8adc1b9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "15138d70e874530448238ae7a36cf1acb39ac5a8691014d56867b035ebb5751f", "generation_seconds_batch": 17.492815589997917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.870182991027832, "target_z": 2.6562498567676704, "ce": 5.935244083404541, "genuineness": 4.13078498840332, "identity": 0.2245745062828064, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6d3cd42552fa31a88aa5a973cdd67c88b067509eac310957a7b6267aa8adc1b9.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "6d771f9a63843265cda029dce0f5a93ee530c46cfb0899c8e1ef9f4030ece42b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "dc36cc1b8cb37f3b2c9959e340f00909845b2129d30e35ca2f5963a97f46af68", "generation_seconds_batch": 23.024417896056548, "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": 6.173524379730225, "genuineness": 4.796044826507568, "identity": 0.5438407063484192, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6d771f9a63843265cda029dce0f5a93ee530c46cfb0899c8e1ef9f4030ece42b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "6e3cfccecf65b3685873d613a0dbaecde2c64d16255eda7979e40b7362e187c6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "87f8e113a042a092ecb2e041939b78c3c691b4fd32ac22831b34f8cc653bff82", "generation_seconds_batch": 24.165765804005787, "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": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.6497112512588501, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6e3cfccecf65b3685873d613a0dbaecde2c64d16255eda7979e40b7362e187c6.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": "6e800562c57ca0f1dd4193877e68205a9c86d971b060704752f8f98e4035097a", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb64a3997a53d0b0e89b2adea235bb0216d94f156500b8d2b0538371d66e5621", "generation_seconds_batch": 17.47242779203225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 6.046441555023193, "genuineness": 4.406930446624756, "identity": 0.7308398485183716, "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_high/6e800562c57ca0f1dd4193877e68205a9c86d971b060704752f8f98e4035097a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "6f331713180e59d3ccbd1d6af98b03eff21ff7c63af11555ecf9712a13caec14", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "eb8ef78bbc33433da30f07588a80af51f71bdb03d64fce85b8f65cf0330eb12d", "generation_seconds_batch": 23.666435455088504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3239998817443848, "target_z": 2.0, "ce": 5.681078910827637, "genuineness": 4.118232727050781, "identity": 0.3506987690925598, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6f331713180e59d3ccbd1d6af98b03eff21ff7c63af11555ecf9712a13caec14.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "6f33c952d516ecc5cc6c121a5d4903862955d6f41382638802fffa842d0101fd", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d977a349ff81a0b57ab082063b4aa06cbba5a04de91da5076a569259bce37b7a", "generation_seconds_batch": 17.404954246012494, "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": 6.117925643920898, "genuineness": 4.482243061065674, "identity": 0.8065078258514404, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6f33c952d516ecc5cc6c121a5d4903862955d6f41382638802fffa842d0101fd.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "6feafb6d5625c4fd4b48f8a36c83b80b1ee7b9d9fda737af492228a3c632743b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a93cd1633094a601af1520ab5392c104cf5fcd839ed113d25f6d460d71b6cb8e", "generation_seconds_batch": 22.72342200600542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1484408378601074, "target_z": 1.789062320959588, "ce": 6.221180438995361, "genuineness": 4.143336772918701, "identity": 0.5478628873825073, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/6feafb6d5625c4fd4b48f8a36c83b80b1ee7b9d9fda737af492228a3c632743b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "70805d16aff35d617aacddaf0c266d4d5abeec3dae962bf074e98779d95f8f9e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e436e98c9dbcae55f93383a87d5a2ac627e1339f45e9475e51e497faada47c20", "generation_seconds_batch": 15.984864283003844, "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": 6.221180438995361, "genuineness": 4.7081804275512695, "identity": 0.7151349186897278, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/70805d16aff35d617aacddaf0c266d4d5abeec3dae962bf074e98779d95f8f9e.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "70a2a0b43908624cb96ba522c3edb199fe345ece640b8c6754c39866c422f61f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 129, "seconds_generated": 10.32, "mp3_sha256": "e3373361c48237de6dc254215867f660ed1420e59ecedaa5f7fb39fd93b8f221", "generation_seconds_batch": 21.365153088001534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0723299980163574, "target_z": 0.4960937679040412, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.6247230768203735, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "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_high/70a2a0b43908624cb96ba522c3edb199fe345ece640b8c6754c39866c422f61f.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "70a3f631aa65ea5db0a74526056dceb794edf38bdbd0048e16aa890a706b63a6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "47530e15da80dd189d0c1bc3223144d4ed8c79936f408c2a5d36f5d7e2f9e1ca", "generation_seconds_batch": 16.76070205704309, "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": 6.38003396987915, "genuineness": 4.7458367347717285, "identity": 0.4200228154659271, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/70a3f631aa65ea5db0a74526056dceb794edf38bdbd0048e16aa890a706b63a6.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "723a0a8212dee97031099128066517f142ae9d781d3513f498917a8817531de9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "2c9fd03069f338c1a280ce6cdaeaadaf6ed78d36dea29e5fa447d5153291f41c", "generation_seconds_batch": 17.811416417942382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.871519088745117, "target_z": 3.859375071616165, "ce": 5.788304805755615, "genuineness": 3.8169827461242676, "identity": 0.4918573796749115, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/723a0a8212dee97031099128066517f142ae9d781d3513f498917a8817531de9.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "72865555ff16e19a70f2f63044e7040861083d311dd5a4587d8b0742a1205412", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a07e98e1ee2b05d8e0dc1e8ed1c65bb92185ca94c87c21df479ca47754a336d0", "generation_seconds_batch": 24.426989419036545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7675402164459229, "target_z": 0.1298827901199485, "ce": 6.141753673553467, "genuineness": 4.7081804275512695, "identity": 0.7814399003982544, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/72865555ff16e19a70f2f63044e7040861083d311dd5a4587d8b0742a1205412.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7385e14852b548d15bf9e0828b74784bc14ded7c34b7a1eaca3448f0ecb5be47", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "869d5c82137965ee81c1d63af942de67fed3736d35fa6318909595378332babc", "generation_seconds_batch": 23.876450276002288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.844174385070801, "target_z": 2.625, "ce": 6.0146708488464355, "genuineness": 4.168440818786621, "identity": 0.5259361267089844, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7385e14852b548d15bf9e0828b74784bc14ded7c34b7a1eaca3448f0ecb5be47.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "742e32d01566cd1317d3115c2284b42af8e7fdea6183bfa9e2898a7016322f2a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fc7d78c6069d9509dead24986b11ec42fab2a4fa91d19e36a3a83d2d21c7a515", "generation_seconds_batch": 22.94138723402284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7297463417053223, "target_z": 0.08447269877209788, "ce": 6.260893821716309, "genuineness": 4.3881025314331055, "identity": 0.3900725245475769, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/742e32d01566cd1317d3115c2284b42af8e7fdea6183bfa9e2898a7016322f2a.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "742e3dbcebaec1315f367f85e7ca4462a7c9065c70b451bd462614ff6d45ddbc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b9a495e121b8132cb97acf0432f853adc6a572f716307340a0e5091573e0abad", "generation_seconds_batch": 22.118617647094652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6224603652954102, "target_z": -0.0444335333238609, "ce": 6.284721851348877, "genuineness": 4.570107460021973, "identity": 0.5850324630737305, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/742e3dbcebaec1315f367f85e7ca4462a7c9065c70b451bd462614ff6d45ddbc.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": "744db92497db0fd3c55c361aea1bb63e8bdcb0e30f38cb5f1ac0bcdc15be72bb", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a280d1486a2482cd54d8120539b7a3baabeb5769a5334c92a76d57e268bd57ce", "generation_seconds_batch": 17.412617385969497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3630130290985107, "target_z": 2.046875071616165, "ce": 6.1020402908325195, "genuineness": 4.218649387359619, "identity": 0.27752938866615295, "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_high/744db92497db0fd3c55c361aea1bb63e8bdcb0e30f38cb5f1ac0bcdc15be72bb.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "74ec91cac70ef73406b11d4abbf5654c4361fabe480be5c07e782fc2b14269a5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "8893852708cfc71f3e69014b42566049182f313320f63d76ce7f8a64e18ef615", "generation_seconds_batch": 17.047142937080935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7453057765960693, "target_z": 1.3046875358080825, "ce": 6.157639026641846, "genuineness": 4.582659721374512, "identity": 0.7247916460037231, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/74ec91cac70ef73406b11d4abbf5654c4361fabe480be5c07e782fc2b14269a5.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "74fc817f6c5023f634e2034b88dc3f0e725e4143b65cb06db46b0be402ffb5b0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "705cdf2f40aef1b4c1bf979da14c538c389b1441b65540aa99e37c031be57d17", "generation_seconds_batch": 24.17070268804673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2524757385253906, "target_z": 1.914062320959588, "ce": 6.213237762451172, "genuineness": 4.281409740447998, "identity": 0.6592085361480713, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/74fc817f6c5023f634e2034b88dc3f0e725e4143b65cb06db46b0be402ffb5b0.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "750a8401880e032656778ba70bdccd48aff6dc6581a0a319acb3e5fa4835abf8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b99b5d92e9d34b17469cf3bff532dc3080a7d0181b0252d8e9abe163dfc6a1b2", "generation_seconds_batch": 22.4099025670439, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.221301078796387, "target_z": 3.0781252148484946, "ce": 5.855817794799805, "genuineness": 3.835810661315918, "identity": 0.4639812707901001, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/750a8401880e032656778ba70bdccd48aff6dc6581a0a319acb3e5fa4835abf8.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "75318f5b377c6c9f46780f2c73d14ef37539b3d1811f24236493ecc67cc72d93", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fd7e0e8164c48c22f2f83656ccb4141d142243f5fc9834384982c4b84ab6b7ee", "generation_seconds_batch": 21.553370086941868, "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": 6.070269584655762, "genuineness": 4.758388519287109, "identity": 0.7238059639930725, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/75318f5b377c6c9f46780f2c73d14ef37539b3d1811f24236493ecc67cc72d93.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "75650a5dc8830db7b0d7d1a1d7a689a0705f6d61fedf4bf168c1e05bac1d1338", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "6d10851dd48c08cb8f2336bf2c6696eb8dfbca1496319572b14267d5905c384e", "generation_seconds_batch": 16.050832613022067, "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": 6.348263263702393, "genuineness": 4.6705241203308105, "identity": 0.6729260087013245, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/75650a5dc8830db7b0d7d1a1d7a689a0705f6d61fedf4bf168c1e05bac1d1338.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "758c92efa702cac833730ecbef14db2e68c2aea37e926385ff869bf4df0fd06c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "175a830f9c2ec6331e5db737fd9547e67a3a4fd71f5489b1881bd2411cde8504", "generation_seconds_batch": 22.85636467509903, "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": 5.593709468841553, "genuineness": 4.545003414154053, "identity": 0.647057056427002, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/758c92efa702cac833730ecbef14db2e68c2aea37e926385ff869bf4df0fd06c.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "75a0268ceb550bb8eeecf908211b04e76483509536cffa53206130980f8d489b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "56a92538bb76e206e9807149e3cde76dcbeda911063493079f1c4f05cd808dd9", "generation_seconds_batch": 17.75679139792919, "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": 6.165581703186035, "genuineness": 4.6956281661987305, "identity": 0.5535093545913696, "hypothesis": "Dich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/75a0268ceb550bb8eeecf908211b04e76483509536cffa53206130980f8d489b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "75b7a4a19354fa587d99d18fd42b7288d98318c52b8545f21a56bcb7169a0e16", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e6ead0cb5441f4f458df9b1cfa0e58e2a96c0e13eccbf63430418beec86881d6", "generation_seconds_batch": 24.29668874805793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7927361726760864, "target_z": 0.16015623209595878, "ce": 6.149696350097656, "genuineness": 4.507347106933594, "identity": 0.5343489050865173, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/75b7a4a19354fa587d99d18fd42b7288d98318c52b8545f21a56bcb7169a0e16.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "768413d98ca4abc60aad5bb59a1016be6563e8e1f381ed654274219d061cfec6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5ef60badeae9aa86c86b13b3c5aa74f81497c95b8516c10673b04487bba44a4a", "generation_seconds_batch": 15.907733820029534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6057984828948975, "target_z": -0.06445311604797939, "ce": 6.332377910614014, "genuineness": 4.419482707977295, "identity": 0.3068397343158722, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/768413d98ca4abc60aad5bb59a1016be6563e8e1f381ed654274219d061cfec6.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "769e1799bdbd4c14d175b1c7f8735d6af42be74d0e01b3684f2e66fd5370f43c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f79b1cdaa0e8b21a49d77ad620930e7c8d20b047d9472d9f21a4ad2cf0da626f", "generation_seconds_batch": 24.43296765605919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3684688806533813, "target_z": -0.3496093481439382, "ce": 6.2767791748046875, "genuineness": 4.337893962860107, "identity": 0.6655737161636353, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/769e1799bdbd4c14d175b1c7f8735d6af42be74d0e01b3684f2e66fd5370f43c.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "77228726e69dc22e4bdb145b09cc989e1ba8e9d753cd0b90eca09b8a55ad48ae", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f7588c88b93a51e72a773274a75c4a3f1f3d203f709b07462a438014aacf109d", "generation_seconds_batch": 22.750165203004144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8358131647109985, "target_z": 0.21191409383207213, "ce": 6.022613525390625, "genuineness": 4.6956281661987305, "identity": 0.7977436780929565, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/77228726e69dc22e4bdb145b09cc989e1ba8e9d753cd0b90eca09b8a55ad48ae.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "77331d7e5dcca00ecbfc25b5258e686ea4de1fab45102e1d191d8d7132698414", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f6ce907a6e69c787f5ec08444e6a8be40bef8702924d80b91893ba910ad5a537", "generation_seconds_batch": 21.81757136201486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4039411544799805, "target_z": 0.894531160479794, "ce": 6.062326908111572, "genuineness": 4.381826400756836, "identity": 0.6122941374778748, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/77331d7e5dcca00ecbfc25b5258e686ea4de1fab45102e1d191d8d7132698414.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "3600d31a941727c360cf2c66733a011a3f9804fb23995e02f11f6eb5f499d4e4", "generation_seconds_batch": 16.429206655011512, "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": 6.181467056274414, "genuineness": 4.394378662109375, "identity": 0.6105890274047852, "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_high/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "782715c8857ffbaeb7d9accea07e913d4fe48199e871039abe57130c32ef8893", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c9e49f1bd0a051dda057d63817dfae7b8aca83d76b793103c5715466dd543fab", "generation_seconds_batch": 17.583302023936994, "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": 6.475345611572266, "genuineness": 4.519899368286133, "identity": 0.4663510322570801, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/782715c8857ffbaeb7d9accea07e913d4fe48199e871039abe57130c32ef8893.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "78473f4639bd5cbbe1a658552a9ae500f930bdfc0d2344f243b115375dfbe93f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f1c576d7800a771589e5302be61c555e33f98dd5dd8ea3abfbf960520a1035e7", "generation_seconds_batch": 22.37035262794234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.286322593688965, "target_z": 3.1562498567676704, "ce": 5.609594821929932, "genuineness": 3.6663577556610107, "identity": 0.5398080348968506, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/78473f4639bd5cbbe1a658552a9ae500f930bdfc0d2344f243b115375dfbe93f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "7859492fe04a5f19aa3f34d6cac8536b8bcf7b2c2ab481918795ae753ddf514a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f9c0a79a3e66c81c4a39a9952f2b22329ec21f3d2a1c4fb5c16ff6df49be4610", "generation_seconds_batch": 21.076287613017485, "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": 6.324435234069824, "genuineness": 4.545003414154053, "identity": 0.6051535606384277, "hypothesis": "La porte 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_high/7859492fe04a5f19aa3f34d6cac8536b8bcf7b2c2ab481918795ae753ddf514a.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "78ef31b47a53fd4f94437560111dd732a9e837368f51ff01f759adf42c73ee30", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4d5754639f5a1aaa03f230cb7778b9d60d5c41753017af955029411bee1a35f1", "generation_seconds_batch": 17.42719495401252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6889045238494873, "target_z": 0.03540044321812107, "ce": 6.443574905395508, "genuineness": 4.469690799713135, "identity": 0.4374915361404419, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/78ef31b47a53fd4f94437560111dd732a9e837368f51ff01f759adf42c73ee30.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7919fe07db47db98952acd4e29e0c9bc5932d16b0ce7de0a2dfc922ef51d0977", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1b23e81883c1097b8c786ad72c41541c808ea1a0af8cf0cfb45bad103f7bf393", "generation_seconds_batch": 24.495827788021415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.384434700012207, "target_z": 0.8710937679040412, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.724660336971283, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7919fe07db47db98952acd4e29e0c9bc5932d16b0ce7de0a2dfc922ef51d0977.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "f8952597730b88bb3a7f882548f3256cfee795f3a6576299cb15193830379d49", "generation_seconds_batch": 16.866215236950666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.0305562019348145, "genuineness": 4.080576419830322, "identity": 0.567582368850708, "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_high/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "7972e4de08b1c402948c09c6c4dfb88add37000d01f275031d78f2c625f6a82b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4e9c19b4a51f653e2b9a62140b27ef4fc5e3b2a87455e392e858ac27d19535d8", "generation_seconds_batch": 24.001239183009602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.641270875930786, "target_z": 1.1796875358080825, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.4944699704647064, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7972e4de08b1c402948c09c6c4dfb88add37000d01f275031d78f2c625f6a82b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "79ba396306752064920989eb0502f060e22ce415a02aa4714f9911309bf2aacd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9d13093dc608c168bc424492835d9770cd36a0566530c09e1335dee1399f00f2", "generation_seconds_batch": 23.057995185954496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.057410478591919, "target_z": 1.6796875358080825, "ce": 6.006728172302246, "genuineness": 4.344170093536377, "identity": 0.4766657054424286, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/79ba396306752064920989eb0502f060e22ce415a02aa4714f9911309bf2aacd.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "79ba7c7e3b992c04de706fae42360000b054e7fab75847e1d3775460e9fa2f59", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "908833c7b4daf91a2df9343889d242340e1aed9ea55552fe2baf01d20255f9b4", "generation_seconds_batch": 23.450546634034254, "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": 6.165581703186035, "genuineness": 4.6203155517578125, "identity": 0.725680947303772, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/79ba7c7e3b992c04de706fae42360000b054e7fab75847e1d3775460e9fa2f59.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "79e24c57f92a550e7565fafd92bdcd39a25fda4c4517f321015c3ae341d23813", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "20763e2b44aa60d848df5c3efd81b3b0084631786872a4442e08fff237cfa94a", "generation_seconds_batch": 18.166802349034697, "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": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.3849750757217407, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/79e24c57f92a550e7565fafd92bdcd39a25fda4c4517f321015c3ae341d23813.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "7a186bd006757adf487bcb9e510de284798aed4e7da44db4265aaa825e26edad", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9b3426d4483a5258cc9812e9b24d32cfc227521f90d81bdd47dbc371b775ac72", "generation_seconds_batch": 15.514108174014837, "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": 6.165581703186035, "genuineness": 4.582659721374512, "identity": 0.6167075634002686, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7a186bd006757adf487bcb9e510de284798aed4e7da44db4265aaa825e26edad.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7b168c3efe00231d57ca0c100d341109c90f0586831beb8c744d98f0608380c7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bae962e756deeb3814574381cc8ac8cd72d84393c0c1e4c8c15d20a1fe4a1638", "generation_seconds_batch": 17.65382483799476, "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": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.7085311412811279, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7b168c3efe00231d57ca0c100d341109c90f0586831beb8c744d98f0608380c7.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "7b5fb97993a127c9925dd639c34ec69427c10083ba8b02a74169bb3231fd55c4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "08aff7612914c51ec43c2970723b487562118d4a26efd2829a3164e63f4ab467", "generation_seconds_batch": 24.653747462900355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8168296813964844, "target_z": 1.3906249283838352, "ce": 5.982900142669678, "genuineness": 4.344170093536377, "identity": 0.6078245043754578, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7b5fb97993a127c9925dd639c34ec69427c10083ba8b02a74169bb3231fd55c4.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "7b7d6d585884a969372562132d981b7edef3f451a55a761f8647f8b9bae2267a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "20ff6bf6952f5bdc5e7aa402aedfcef8fc461b5d19857c6f7df0be3da90aa368", "generation_seconds_batch": 17.084808100946248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9390352964401245, "target_z": 0.33593753580808244, "ce": 6.117925643920898, "genuineness": 4.720732688903809, "identity": 0.6147359013557434, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7b7d6d585884a969372562132d981b7edef3f451a55a761f8647f8b9bae2267a.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "7bb1b3776a1c54582720b33363e3ceaa25b4837c15685c8334ba2dfe3d799436", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "450e1f77d058c1862c1c41fcae0682d6d2598682f334d1ab3fc01b76fdae4532", "generation_seconds_batch": 22.11060465197079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8089916706085205, "target_z": 0.17968753580808242, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.5992752909660339, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7bb1b3776a1c54582720b33363e3ceaa25b4837c15685c8334ba2dfe3d799436.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "7c64f487814388e5fc3b7941b21c5412450ee85864abb9c6f00fa0302f3400da", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f1bb9d578d2b20e4dcf9ce9bebd91ada5b252eb95847db01cbc53687e73e1df9", "generation_seconds_batch": 16.818530895048752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7015024423599243, "target_z": 0.05053709258996136, "ce": 5.998785495758057, "genuineness": 4.143336772918701, "identity": 0.606885552406311, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7c64f487814388e5fc3b7941b21c5412450ee85864abb9c6f00fa0302f3400da.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "7ca296d2363df27d03cec6c2c6c1b84b83c402ba761cbf5b52f7cd1387e47793", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5bd2a1e4b03a1c17f547582a1d1683ac326d1364c689b72bc732f6e4c4e74054", "generation_seconds_batch": 22.035793457063846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.454043388366699, "target_z": 2.1562498567676704, "ce": 5.788304805755615, "genuineness": 4.049196243286133, "identity": 0.6319308876991272, "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_high/7ca296d2363df27d03cec6c2c6c1b84b83c402ba761cbf5b52f7cd1387e47793.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": "7cbf7a84aadb7677482fb599bdec2f20e874b9924ae1a81b5ae4fcc19a114b67", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "26a3d31499bf3217e38960e61996aa6e56e8cd466f970c102d450dc3b370a630", "generation_seconds_batch": 23.107930066995323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2979912757873535, "target_z": 1.9687501432323298, "ce": 6.205295085906982, "genuineness": 3.9550557136535645, "identity": 0.5659921169281006, "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": 0.120000000000001, "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_high/7cbf7a84aadb7677482fb599bdec2f20e874b9924ae1a81b5ae4fcc19a114b67.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "7cdbb235f1668980c5749e39b93e8732fcc33642f788c60b96541539777b96bd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e6857c512f26215dcd6b71ab51a23eadcf79d0ae51b8460a6adaa96eafffd643", "generation_seconds_batch": 24.653747462900355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9114010334014893, "target_z": 0.302734419760103, "ce": 6.348263263702393, "genuineness": 4.168440818786621, "identity": 0.44263121485710144, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7cdbb235f1668980c5749e39b93e8732fcc33642f788c60b96541539777b96bd.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7cfd88e9e314098f2df5b83863afa3fa3a27b47fa67d209900493491cd7017cf", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2cd965c80b3d59ca8bfabc84bbdf730353aa3b755960bbacafb9518a4ae957ec", "generation_seconds_batch": 23.713507132953964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2589781284332275, "target_z": 1.9218750716161648, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.6507853269577026, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7cfd88e9e314098f2df5b83863afa3fa3a27b47fa67d209900493491cd7017cf.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7d3f1e38d8ede26fde9ed4e74bbd3de4b77ce16b03ce011c139bdf525cec56fe", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9eb241228e9ca6ef5f572b483eae47aeaedcd2fdf2a18181fab3ded5f4d0e665", "generation_seconds_batch": 18.40923382504843, "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": 5.974957466125488, "genuineness": 4.394378662109375, "identity": 0.5398346185684204, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7d3f1e38d8ede26fde9ed4e74bbd3de4b77ce16b03ce011c139bdf525cec56fe.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7d45bdb7f8e85731358958e3eb301a99c37d154902cae60653e455f8b1002da5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "34e80aed64134d4a72975d7ab3b45503046ab154b16a872341f681b88e66ecb8", "generation_seconds_batch": 23.093001386034302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.169283390045166, "target_z": 3.015624928383835, "ce": 5.919358730316162, "genuineness": 4.036643981933594, "identity": 0.5020791292190552, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7d45bdb7f8e85731358958e3eb301a99c37d154902cae60653e455f8b1002da5.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "7d7524ebd7a35a051dcc8d00bbe3a4f16f9ea150b60afa1cfbe4e2efb84d7455", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "13fc5911ab7ddedd65c8329ecf6278582e990a9d4de437e4a3170286a68d764a", "generation_seconds_batch": 23.876450276002288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.6483578681945801, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7d7524ebd7a35a051dcc8d00bbe3a4f16f9ea150b60afa1cfbe4e2efb84d7455.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "7df085a549fa71e6926d115c9c26e680742f0aefeb8a96bcbce9c407edc0c209", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d1d7774c982e9a3900381de6fc01f5760fa28bdf8f0281e3507f966a5c15735b", "generation_seconds_batch": 17.964187899022363, "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": 6.157639026641846, "genuineness": 4.65797233581543, "identity": 0.6280816793441772, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7df085a549fa71e6926d115c9c26e680742f0aefeb8a96bcbce9c407edc0c209.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "49a8d03ba92e2cc06e6271b03b0aff530c9a76159d8c75b1163f543761f3f468", "generation_seconds_batch": 18.90785944298841, "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": 6.38797664642334, "genuineness": 4.532451152801514, "identity": 0.39103224873542786, "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_high/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.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": "7f918b95f21238b92b89b4702f7cbb5be51d5230639e135b8a93e6e08aaee938", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8cfaeac5fb53f023a75d058c00598a52728b3ef09c1591abfa8faffcaeff9c2", "generation_seconds_batch": 18.049854250974022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.429949998855591, "target_z": 0.9257813037121236, "ce": 5.982900142669678, "genuineness": 4.532451152801514, "identity": 0.8756668567657471, "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_high/7f918b95f21238b92b89b4702f7cbb5be51d5230639e135b8a93e6e08aaee938.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "7fcf8f538201522c5451023c1313dd64d7642622a44061469863ea9abae167bc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6c23bc351caf2aae2e0859ae468b3d5104e75dd77ca918934f6571fb1bcebe12", "generation_seconds_batch": 23.93550693197176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532069683074951, "target_z": 2.25, "ce": 4.818802356719971, "genuineness": 3.8671910762786865, "identity": 0.4798324406147003, "hypothesis": "La puerta sigue abierta. He dejado tu abrijo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7fcf8f538201522c5451023c1313dd64d7642622a44061469863ea9abae167bc.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "7fdc507870500842ab1b0bb66365d0934440247946a33964f95c2b1412b0cfa3", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "37eb45fec20be5ae633ec4646296e9433f5d8c7aa0df37c3c57255c1e5d5972f", "generation_seconds_batch": 17.206087893922813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6608638763427734, "target_z": 0.0017090481581468258, "ce": 6.252951145172119, "genuineness": 4.432034492492676, "identity": 0.46057236194610596, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/7fdc507870500842ab1b0bb66365d0934440247946a33964f95c2b1412b0cfa3.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "80240800c38860b3ff6e44408cbc68c1f2a0bd6ec23bedbdaaf947d568af475d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f15952cd74a06527601a898d100c5a4e645abe30b7d0619616f1bb0591b0cea2", "generation_seconds_batch": 22.56464088591747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.758310079574585, "target_z": 1.3203124641919175, "ce": 6.006728172302246, "genuineness": 4.293961524963379, "identity": 0.6298769116401672, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/80240800c38860b3ff6e44408cbc68c1f2a0bd6ec23bedbdaaf947d568af475d.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "805e344a81b0c9462fe3a6b4810cfe86730dedde2d9f524f2f0376f99a086c1e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4a216e1d23459ca44074e0300fb5874cec77737bac3a988d484dc7193e101e20", "generation_seconds_batch": 17.28753367101308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3239998817443848, "target_z": 2.0, "ce": 5.927301406860352, "genuineness": 4.413206577301025, "identity": 0.49660345911979675, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/805e344a81b0c9462fe3a6b4810cfe86730dedde2d9f524f2f0376f99a086c1e.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "806250d7db9bf109a4f78cb72a4d075922cce835359bf29846f6e42a0eef37ea", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f80363f02325cfe8a58727702cd1e62d8b89c4dde8dc4043fcfd054a35aa4a3e", "generation_seconds_batch": 18.739263223018497, "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": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.647711455821991, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/806250d7db9bf109a4f78cb72a4d075922cce835359bf29846f6e42a0eef37ea.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "8075ca37fe7f3495c0581e658bd2c65db28d0a54e71ae46ffa79d26780e0a5dc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "234ef7e507e59be17c4d0bd8d91857dfa5f8a6ffb28bcf75e17a8718ea149483", "generation_seconds_batch": 22.62674724601675, "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": 6.0305562019348145, "genuineness": 4.444586753845215, "identity": 0.6435182094573975, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8075ca37fe7f3495c0581e658bd2c65db28d0a54e71ae46ffa79d26780e0a5dc.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": "807914549bb87395703f9ba967f20b7a97a05d758807baacf46168e795ea8c74", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab1e05a53b711c448fd070591af79e5502cdfa10a864f2153cd659211d376aff", "generation_seconds_batch": 17.971176479943097, "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": 6.260893821716309, "genuineness": 4.193545341491699, "identity": 0.6840686798095703, "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_high/807914549bb87395703f9ba967f20b7a97a05d758807baacf46168e795ea8c74.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "80a435e1d2c0fef98b314ae5f399850b6cdb1115bb307ee6f9933496b8c58d5d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "5748461abba9a87af41a7c4496e54a48db492f50e6143367aaa74adaf7060f98", "generation_seconds_batch": 23.024960463983007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.979384422302246, "target_z": 1.585937679040412, "ce": 6.268836498260498, "genuineness": 4.256305694580078, "identity": 0.6190400719642639, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/80a435e1d2c0fef98b314ae5f399850b6cdb1115bb307ee6f9933496b8c58d5d.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "612c64efdcfc01a3e7420e0bed9f0ab6efb77e5d02b7ac61d7e5988aca562c0f", "generation_seconds_batch": 17.715684288996272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4104433059692383, "target_z": 0.9023436246717115, "ce": 6.340320587158203, "genuineness": 4.507347106933594, "identity": 0.6453942060470581, "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_high/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "80fd9b55695e46f9381155961ffe4953ff49590ebc0018fd80e4ade743c15915", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "994929b95b39595ce521fa3c52691e1337b5376b070fe9b972720e4077839e21", "generation_seconds_batch": 23.246912455069833, "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": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.36330339312553406, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/80fd9b55695e46f9381155961ffe4953ff49590ebc0018fd80e4ade743c15915.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "81636ac04536a94a662db45437f6861f817cb8bd1e7eabfcae34d0759b6fd9a0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fcaedd6f7f45ea10cbdc263c5f54d97440fa150aefa17714c950d083c4635eef", "generation_seconds_batch": 23.904840764938854, "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": 6.3641486167907715, "genuineness": 4.494794845581055, "identity": 0.5936983227729797, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/81636ac04536a94a662db45437f6861f817cb8bd1e7eabfcae34d0759b6fd9a0.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": "81e074779de8004541c3941a16eb3922f6a9868345dd538f37a51da6313073ec", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02e507e6e9d39703e7fd86d470effc6f774b6bd35bed4e81118d2e7796d60fb4", "generation_seconds_batch": 18.53081536490936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.753143787384033, "target_z": 2.515624928383835, "ce": 6.117925643920898, "genuineness": 4.231201171875, "identity": 0.11085259914398193, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon. 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_high/81e074779de8004541c3941a16eb3922f6a9868345dd538f37a51da6313073ec.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": "829c49d654be6d36113751dc628ff8eebd08b6ba8bae8eb863157646ffb367ab", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1a77074a4c05a181a46b2de546c9b4053b1af168c6c4c6bc3a6cc08c309e66f5", "generation_seconds_batch": 22.34890999691561, "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": 6.348263263702393, "genuineness": 4.042920112609863, "identity": 0.38870614767074585, "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": 0.120000000000001, "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_high/829c49d654be6d36113751dc628ff8eebd08b6ba8bae8eb863157646ffb367ab.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "82a58569a5ae6f9ae63bcddbcc6c52cdc9a5c61aa0489f37cf405accce232236", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "307bb627c0413b58eae996dc0f8ad4f672889546fab3965947861087b62b66c3", "generation_seconds_batch": 22.4099025670439, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15627908706665, "target_z": 3.0, "ce": 6.23706579208374, "genuineness": 4.1998209953308105, "identity": 0.3201085329055786, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/82a58569a5ae6f9ae63bcddbcc6c52cdc9a5c61aa0489f37cf405accce232236.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "82cb3cdcfbceb59b9fccaa8f02c27b7ab130bd3f65b6c929056a51d8af8499a7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ba3861b9b888959b646d0df7ee02b8b21984a511f4af54ab2115589983f8c8c6", "generation_seconds_batch": 23.673339955043048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.024899482727051, "target_z": 1.6406249283838352, "ce": 6.1020402908325195, "genuineness": 4.15588903427124, "identity": 0.7092307209968567, "hypothesis": "Ik heb een klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/82cb3cdcfbceb59b9fccaa8f02c27b7ab130bd3f65b6c929056a51d8af8499a7.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": "82ddfbfc592d8dfa848c5502a47866ed7c656f1bd36e57b6275d60d35763671b", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6dcc15c1c4cf813c82f37205d90e8f5db058b2938e10e581308b4bd7ad1c282", "generation_seconds_batch": 18.049854250974022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4722142219543457, "target_z": 0.9765626074242473, "ce": 6.070269584655762, "genuineness": 4.607763767242432, "identity": 0.7850502729415894, "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_high/82ddfbfc592d8dfa848c5502a47866ed7c656f1bd36e57b6275d60d35763671b.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "834877cfcbd812fb98cf4c193ffc4d35a245fbed7ab3d365091b203be3ff8cbc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "418ddd6cec7c8b1f8eaecde57eb5f2058a782d6f2526b22916ad78418270a4e5", "generation_seconds_batch": 23.25120004697237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7713146209716797, "target_z": 1.335937679040412, "ce": 6.221180438995361, "genuineness": 4.2374773025512695, "identity": 0.7048178315162659, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/834877cfcbd812fb98cf4c193ffc4d35a245fbed7ab3d365091b203be3ff8cbc.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "835db777e6d7acf1c30bc941d852a42f37bf5986ed6df43530c8b3e17f47f7b4", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a566c40e84392fa1596c62a2aaf8589c1aaca23db6114f285c3248816fcac858", "generation_seconds_batch": 16.891833167988807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6607621908187866, "target_z": 0.0015868709809008282, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.6670318841934204, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/835db777e6d7acf1c30bc941d852a42f37bf5986ed6df43530c8b3e17f47f7b4.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "836ddd01defa3688cdaad37e28dea4ad4b4add823fe5a0ca0886e18822cee854", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b08079c854fb63f7c254e54c4dea970867133b48332ace79d4298c24d6d846fb", "generation_seconds_batch": 24.96598624100443, "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": 6.0146708488464355, "genuineness": 4.6705241203308105, "identity": 0.6053333282470703, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/836ddd01defa3688cdaad37e28dea4ad4b4add823fe5a0ca0886e18822cee854.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "83959ee8a3e3c393d1bedc11814896ad8cee14ef0fa6a8d85dcbb29a9d372581", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0e412965898e199932125dddc9e779762307e649aa6a3dfab78e844e58d015f6", "generation_seconds_batch": 15.745475627947599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.576249122619629, "target_z": 1.1015626074242473, "ce": 6.24500846862793, "genuineness": 4.030367851257324, "identity": 0.2722247242927551, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/83959ee8a3e3c393d1bedc11814896ad8cee14ef0fa6a8d85dcbb29a9d372581.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "83c8e964cb464f49b2ca8457c1882af07b0bf0a9008c494b6aa41edd65d45ecb", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0602ecee246127592e065817a60bf3e6db4e1a4f837bfa83fd45f536b3d64d29", "generation_seconds_batch": 16.66556944604963, "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": 5.982900142669678, "genuineness": 4.632867813110352, "identity": 0.7813495993614197, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/83c8e964cb464f49b2ca8457c1882af07b0bf0a9008c494b6aa41edd65d45ecb.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "843603920f7460262c02a81e9d91c408c4d48e3d3326ee777650f0af622e6b35", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "685e5170bf67ec51086b388569bf82eeaaab75a29a54e3b37f1db833d9a91b6d", "generation_seconds_batch": 22.637672243989073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.861009120941162, "target_z": 0.24218753580808242, "ce": 6.173524379730225, "genuineness": 4.268857479095459, "identity": 0.7447633743286133, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/843603920f7460262c02a81e9d91c408c4d48e3d3326ee777650f0af622e6b35.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": "847b4ff435e8a898c96556d5a41dc8fae58a5675c09452dee079c5a2a0d760b6", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9aea86b69ce8f8430263425a17a9be0119c308684e83ed8fec505c4ec950fd2", "generation_seconds_batch": 17.142593416036107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1939563751220703, "target_z": 1.8437501432323298, "ce": 6.173524379730225, "genuineness": 4.300237655639648, "identity": 0.5483094453811646, "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_high/847b4ff435e8a898c96556d5a41dc8fae58a5675c09452dee079c5a2a0d760b6.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": "848c234a08061f1c4fe7bb07bdbe432a04d63c0cb44b03aa2e83a039e618ea33", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ac69012b2d3844addc2a35599ab0d0b311242a5c21b9b51aa2183eb6fcd63fe", "generation_seconds_batch": 25.696245218045078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.024899482727051, "target_z": 1.6406249283838352, "ce": 6.252951145172119, "genuineness": 4.325342178344727, "identity": 0.6623073816299438, "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_high/848c234a08061f1c4fe7bb07bdbe432a04d63c0cb44b03aa2e83a039e618ea33.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "84b085bd85dc01ba84e445c211234604b6714f6018e5ba15968ace5a553a0ee2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b0157599f3a2a2151b35c2b51ff673e3117e1d37aa7c72237a92f54924644390", "generation_seconds_batch": 24.377222332987003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.834187626838684, "target_z": 0.20996097778409273, "ce": 6.300607204437256, "genuineness": 4.720732688903809, "identity": 0.5155311226844788, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/84b085bd85dc01ba84e445c211234604b6714f6018e5ba15968ace5a553a0ee2.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "84e2d0e873564d9a8212c5aa70f44d2cb2e61bd76261f112e656a1b95d988895", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6d2ee792cba3c4c865b807058df4c6b540443696d8431fa821ff2399d592a037", "generation_seconds_batch": 24.371148369973525, "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": 6.09409761428833, "genuineness": 4.469690799713135, "identity": 0.6854777932167053, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/84e2d0e873564d9a8212c5aa70f44d2cb2e61bd76261f112e656a1b95d988895.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": "851e3e0aa6e75c768741a1a0a3376d5c231c7eb23f9f68164126c62e3a9953e8", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6dde24da5257e758ebd3c6bc84630112aa8ad2f1f56fc6527153413e06b1fe66", "generation_seconds_batch": 17.855276351911016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6607773303985596, "target_z": 1.2031249283838352, "ce": 5.919358730316162, "genuineness": 4.482243061065674, "identity": 0.653567910194397, "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_high/851e3e0aa6e75c768741a1a0a3376d5c231c7eb23f9f68164126c62e3a9953e8.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "85296e696bbaf014e007a198c0b6aa2f4db87f41b89a2ba1150094a5d1e402aa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e5f33d8e861139d3e364b69d1b3d761c7a9191ec717ab9d6f716d81a130930fa", "generation_seconds_batch": 25.53965748997871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8114299774169922, "target_z": 0.1826172098800515, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.45026323199272156, "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_high/85296e696bbaf014e007a198c0b6aa2f4db87f41b89a2ba1150094a5d1e402aa.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "853c7d0e39246b6b8a8b972cc7dac97acefaa423c8d292c7c4f7c65294542d14", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5b8a89540a8963ae0b6bb59116b1cf9ee2f3015fb3e145a17f1d5702aeb51db3", "generation_seconds_batch": 17.166892051929608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.163360595703125, "target_z": 0.605468839520206, "ce": 6.046441555023193, "genuineness": 4.595211505889893, "identity": 0.7419497966766357, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/853c7d0e39246b6b8a8b972cc7dac97acefaa423c8d292c7c4f7c65294542d14.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": "855b667b39c16a63446e9587aea9aca0dc628fc90b821f0c3b90ed3888cd05ea", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fec19d2c00050f6ab610488bac1431b230d47d72dc68bc9508f636d77cc0b9a4", "generation_seconds_batch": 17.026644231984392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 5.831989288330078, "genuineness": 3.9927117824554443, "identity": 0.34001898765563965, "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_high/855b667b39c16a63446e9587aea9aca0dc628fc90b821f0c3b90ed3888cd05ea.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "8563dfcadc0379a2c77399076749fe9142c39793e89d862374879e6499f1488b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2ef67c50e43dc3a9dfb6cd7f33422f1fc577e89a0ff8333c4f3d5b4591e054ba", "generation_seconds_batch": 24.844639451010153, "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": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.6779725551605225, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8563dfcadc0379a2c77399076749fe9142c39793e89d862374879e6499f1488b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "858baf7ef2aa1f15355274b635227c609b275fcbdcace40cf1c1c71b3399c51b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "db5909e1fcbda2cecec03c5bb141e61a294d2de555e0549618a90ca22eda2683", "generation_seconds_batch": 22.67225395492278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 6.109982967376709, "genuineness": 4.137060642242432, "identity": 0.3880174160003662, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/858baf7ef2aa1f15355274b635227c609b275fcbdcace40cf1c1c71b3399c51b.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "8615b4a59f1f17a416cd81266055af923ecf3a71ee66c395e4a5edf63c754760", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "71b8a319cbfdeec0ca6f8c2ec13ccc6731cb2d2b9f0b4a0a9012f0f5a182a795", "generation_seconds_batch": 22.101699391030706, "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": 6.141753673553467, "genuineness": 4.030367851257324, "identity": 0.19442173838615417, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8615b4a59f1f17a416cd81266055af923ecf3a71ee66c395e4a5edf63c754760.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "8669a40ec3653d7aed587354a8b115f30d05db818f797d97f9bea9c2845676bc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ac0e29353787dfe31edaa076c39cd0cf5a6792f6a68e06cbe23057a5753a208e", "generation_seconds_batch": 23.283530917018652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4104433059692383, "target_z": 0.9023436246717115, "ce": 5.959072113037109, "genuineness": 4.720732688903809, "identity": 0.693669855594635, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8669a40ec3653d7aed587354a8b115f30d05db818f797d97f9bea9c2845676bc.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": "86c3edcadd02c14354b434c2a228973cf5b915c7713faa23db5e0c3ca703fdf1", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d1776722e4ac10b3420dc8892fe97330c00b8dd4e908e7aba929e3f77d816a5", "generation_seconds_batch": 24.911018306971528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.122432231903076, "target_z": 1.7578124641919175, "ce": 5.000987529754639, "genuineness": 4.105680465698242, "identity": 0.6974818110466003, "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_high/86c3edcadd02c14354b434c2a228973cf5b915c7713faa23db5e0c3ca703fdf1.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "86d2952a95fae174cd0123d6ee74da296bcc68766544f44d784ce0ecff8731e7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e69a2826f8701e7f8046a8f74c81a33fa832e2532c3d6e8b2d1add38eb9cd57e", "generation_seconds_batch": 16.47344362095464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892534255981445, "target_z": 1.1171875358080825, "ce": 6.125868320465088, "genuineness": 4.7458367347717285, "identity": 0.7529086470603943, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/86d2952a95fae174cd0123d6ee74da296bcc68766544f44d784ce0ecff8731e7.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "86e0cf447d47d2aaf6c5e12851712f9db6138fca7d1004dd5ebfbff7d01bf52f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e08119520ffef41ac33aac669da36c88a68b49a1c63a26bea8d6279f2887fc5c", "generation_seconds_batch": 17.047142937080935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2914891242980957, "target_z": 1.960937679040412, "ce": 6.038498878479004, "genuineness": 4.494794845581055, "identity": 0.6337387561798096, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/86e0cf447d47d2aaf6c5e12851712f9db6138fca7d1004dd5ebfbff7d01bf52f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "876ce94722d1dc460ad69de97e98eb9c81e9f4404e89379e50246efd5927a965", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e145ec72145077a86b2c10382f367545fff03e08382032215bd1ec1358fc6029", "generation_seconds_batch": 17.48771045799367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9227797985076904, "target_z": 0.3164062320959588, "ce": 6.062326908111572, "genuineness": 4.6705241203308105, "identity": 0.7720234990119934, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/876ce94722d1dc460ad69de97e98eb9c81e9f4404e89379e50246efd5927a965.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "888bceb8953dcb6e011e73966060f6c5fe125477f56c60e3930204613f2c2e9a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6987503cc3a9b1da599bcdeb1d093197c8d6c192cc0c1973d69314e77ea94dbc", "generation_seconds_batch": 23.298010800033808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.844174385070801, "target_z": 2.625, "ce": 5.967014789581299, "genuineness": 3.8671910762786865, "identity": 0.416358083486557, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/888bceb8953dcb6e011e73966060f6c5fe125477f56c60e3930204613f2c2e9a.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "88c7652caf335b72a9cac36dc53da9af01c49645d70d26038aa9d7cfd1d0c689", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b9780f2c1504ff2f43119fc11d5b31f8d11089782ae1b19df1572fff62a7f35f", "generation_seconds_batch": 18.013003390049562, "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": 6.348263263702393, "genuineness": 4.2751336097717285, "identity": 0.29294896125793457, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/88c7652caf335b72a9cac36dc53da9af01c49645d70d26038aa9d7cfd1d0c689.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "89640b0a4cdd62a6141445d8e5260454418c705d5bebdf0e602822217656b265", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c6631135b042259bd1289f1cc70d5876ba5164b989e866cea7f2578ed094e22c", "generation_seconds_batch": 15.3653919310309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8168296813964844, "target_z": 1.3906249283838352, "ce": 6.181467056274414, "genuineness": 4.344170093536377, "identity": 0.48387524485588074, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/89640b0a4cdd62a6141445d8e5260454418c705d5bebdf0e602822217656b265.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "897dd280c0108ce64089dfc92771749b767ec15018073c2f47dd1a2e4994e0af", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f5e2249feb7b63b2f5d80422ee48d4d328363579573b306bfed82faba6de3a7", "generation_seconds_batch": 17.195034984964877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.870182991027832, "target_z": 2.6562498567676704, "ce": 5.728734970092773, "genuineness": 4.168440818786621, "identity": 0.6395710706710815, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/897dd280c0108ce64089dfc92771749b767ec15018073c2f47dd1a2e4994e0af.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "89acfa3817acf79fedbf1513049f4c166d5051544a299cdbdc126447c2c5cf12", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f5ef4d48e1f9292a3d2c943856527a6d07e4971436c3e02b8e1ea8d04b6dcadf", "generation_seconds_batch": 22.538021008018404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5970914363861084, "target_z": 2.328124928383835, "ce": 5.414999485015869, "genuineness": 3.8295347690582275, "identity": 0.47252124547958374, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/89acfa3817acf79fedbf1513049f4c166d5051544a299cdbdc126447c2c5cf12.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "89e3d15cff5333fd649ebb994addae63b4788766e30e8dbaf93648094a56be95", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bf898faa48260b0c61bac7d5e539115538f7f7c01bf816378835edd86244252b", "generation_seconds_batch": 16.59198665106669, "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": 6.1020402908325195, "genuineness": 4.720732688903809, "identity": 0.5642560720443726, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/89e3d15cff5333fd649ebb994addae63b4788766e30e8dbaf93648094a56be95.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "89f71c5125c4627f7e59e05e88ebfe45fc4115cd507c238a9345300b2613c01f", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0bc753efebc939d356a1a1a3690e8da977aaf4cf2ad493d4eeb9c53d898b7893", "generation_seconds_batch": 18.754759349976666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8187448978424072, "target_z": 0.19140623209595878, "ce": 6.1894097328186035, "genuineness": 4.720732688903809, "identity": 0.7489759922027588, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/89f71c5125c4627f7e59e05e88ebfe45fc4115cd507c238a9345300b2613c01f.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "8a6f2868180ed0d3b84aab29a70f9f57ce59b1ca8e92dd9a9b21f822751a6322", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "323b3e60c32f255b9fb22b0c04ee21ca6d8ce4a893dfdd06724b44cbdf99872b", "generation_seconds_batch": 17.583302023936994, "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": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.38122326135635376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8a6f2868180ed0d3b84aab29a70f9f57ce59b1ca8e92dd9a9b21f822751a6322.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "8a901f1e569f05a4955aca8485ca8824b6199cedb46aa68330c5e0df92c42d2c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "aae314c3995b22a3f3432f16b9a2fe623b6a679ab583be70f4556358c2e0c986", "generation_seconds_batch": 23.961544070974924, "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": 6.205295085906982, "genuineness": 4.532451152801514, "identity": 0.7445478439331055, "hypothesis": "Ik ha reeds, dus ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.2222222222222222, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8a901f1e569f05a4955aca8485ca8824b6199cedb46aa68330c5e0df92c42d2c.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "8b0d10e778c0b5d86152aaf4348b3da994736a6f0aaeaa9468d27e59e9f32ec2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "975fc5b11b9c2b94d7b4b3d0c4e340a52d0b248174c5e6e91a42a94e5f40ba26", "generation_seconds_batch": 21.88348264200613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.273318290710449, "target_z": 3.140624928383835, "ce": 5.871703147888184, "genuineness": 3.9173994064331055, "identity": 0.4198530316352844, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude, oude ramen. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8b0d10e778c0b5d86152aaf4348b3da994736a6f0aaeaa9468d27e59e9f32ec2.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": "8baf4243ebc60b561a78f8427c9e091fdede8eeaba34ed21b01345c5500ad9cf", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bff965324398ba43dc683e8085ebce6f8933cc319ca9964e8dbb165ff5ce53ed", "generation_seconds_batch": 18.935195220052265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.271982431411743, "target_z": 1.9375, "ce": 6.252951145172119, "genuineness": 4.250029563903809, "identity": 0.537712037563324, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8baf4243ebc60b561a78f8427c9e091fdede8eeaba34ed21b01345c5500ad9cf.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "8bdd8ef0ad1eff33223b25c190d30165839a0eca70b3d1e19c33384e2cbf4bab", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1eff7a6318e8e20622926445ae6534339f13c586973cf7896c2244fd44f86ff2", "generation_seconds_batch": 22.11060465197079, "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": 6.308549880981445, "genuineness": 4.68307638168335, "identity": 0.6195656061172485, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8bdd8ef0ad1eff33223b25c190d30165839a0eca70b3d1e19c33384e2cbf4bab.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "8c0b81aded3d7a0d2bb0dc7dd437a0e93597c99fb58ed5ae7be897f959a2273a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f8095682262b9ca6dd73d5fb763ea6866d0931526077a47d0a2ef4ec61ff9d1b", "generation_seconds_batch": 17.283323684940115, "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": 6.078212261199951, "genuineness": 4.15588903427124, "identity": 0.6582359075546265, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8c0b81aded3d7a0d2bb0dc7dd437a0e93597c99fb58ed5ae7be897f959a2273a.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "8d6628b122939beb9b1561e84827d50ab0a61e953f4c077e9eaf3417fd115e58", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0aa86cd199f50e6ebaedae073a829a194c4f61a846b0a44f420687b1431cc23d", "generation_seconds_batch": 16.943611936061643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.057410478591919, "target_z": 1.6796875358080825, "ce": 6.356205940246582, "genuineness": 4.406930446624756, "identity": 0.5340742468833923, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8d6628b122939beb9b1561e84827d50ab0a61e953f4c077e9eaf3417fd115e58.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "8dc79b51bf4fa95279549ccd9941ce30d744689c6e655fb70f47fe3c6d083582", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "de97891117c795f0fbe1689941492d6f71613b2d1e084f928fd3a806207ba6a9", "generation_seconds_batch": 23.048965709051117, "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": 5.760505676269531, "genuineness": 4.093128204345703, "identity": 0.5997840762138367, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8dc79b51bf4fa95279549ccd9941ce30d744689c6e655fb70f47fe3c6d083582.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "8e0a2031fd00e5a2b6c5ae6777c4e4ae9d376b800386ac17f772055b855e81da", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1ed58ad42b9d89afeae63c70256408a7a8bc30cc8b89a685098c504f74812113", "generation_seconds_batch": 23.1312564809341, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.026235580444336, "target_z": 2.8437501432323296, "ce": 6.062326908111572, "genuineness": 4.162164688110352, "identity": 0.7258914709091187, "hypothesis": "Ik heb ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8e0a2031fd00e5a2b6c5ae6777c4e4ae9d376b800386ac17f772055b855e81da.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "8e88253985005196ff832ee81275a9460f5dc0908daa16b882c6fa03896ff59e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1cfa7182c47725a7f87e1eaed7a667310fb76fd0243e9372bc6483a4c30807ac", "generation_seconds_batch": 17.47234323504381, "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": 6.372091293334961, "genuineness": 4.557555198669434, "identity": 0.6524765491485596, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8e88253985005196ff832ee81275a9460f5dc0908daa16b882c6fa03896ff59e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "8e9550aa49ae6569cf6ab8e7f822ade809059a4aab3ab281c0d4bd435078e6d8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e23612d95b8efbb5b23b4c8f3eb33a13917569111560170c83b5c35fd4a4d764", "generation_seconds_batch": 22.118617647094652, "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": 6.348263263702393, "genuineness": 4.256305694580078, "identity": 0.36961477994918823, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8e9550aa49ae6569cf6ab8e7f822ade809059a4aab3ab281c0d4bd435078e6d8.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "8eb24e6efee80ab9b5a1e3f9f6fb02ba208e1dd94da6c15a5a5e538ef5d5f87d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a65d17fc77703650afc22318c1d34fc36aa8d0d0b02e2134562f17f96484640b", "generation_seconds_batch": 23.46593542804476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.316161632537842, "target_z": 0.7890623209595878, "ce": 6.229123115539551, "genuineness": 4.193545341491699, "identity": 0.7250268459320068, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8eb24e6efee80ab9b5a1e3f9f6fb02ba208e1dd94da6c15a5a5e538ef5d5f87d.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "8ed01599fae32405f8a99b1787044ef6a671ff8ab36d9d02cf773f22d8b58005", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b892c137a7198d5d01dcaa3461f018882cbb47f847beaa0dd0a00ec78dfbbc6a", "generation_seconds_batch": 22.93818044802174, "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": 5.363371849060059, "genuineness": 4.21237325668335, "identity": 0.46762973070144653, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8ed01599fae32405f8a99b1787044ef6a671ff8ab36d9d02cf773f22d8b58005.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "8f2ac5035f0dec18865bbac44e02bb269eb4f466cdf51240edc91c4cdd10b375", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "764d354f0761f3facf67b811718312f6564ee1200deb6bbce38963767834003b", "generation_seconds_batch": 18.021813014987856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.853694200515747, "target_z": 0.23339851359217517, "ce": 6.133810997009277, "genuineness": 4.6956281661987305, "identity": 0.6455470323562622, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8f2ac5035f0dec18865bbac44e02bb269eb4f466cdf51240edc91c4cdd10b375.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "8f54a7b35a3c475f6243f7799e7f4a4a19e2ec827a2068dbecc476a4d677e79f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e73c9756bc34ac26cec57eb88f5a1f3d6213471b08c8d6641f8f461f79c8e358", "generation_seconds_batch": 23.166388290934265, "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": 5.982900142669678, "genuineness": 4.507347106933594, "identity": 0.4367065727710724, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8f54a7b35a3c475f6243f7799e7f4a4a19e2ec827a2068dbecc476a4d677e79f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "8f9b554e4acf162d6e7f098c98bbc0303af6c5a710be169c14f7e98bff6553f1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "633dfa4c541035b059364f2828bc0b22420d45ff290222e437a4933cef76c4f6", "generation_seconds_batch": 25.47056671709288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.428034782409668, "target_z": 2.125, "ce": 5.240260601043701, "genuineness": 4.331617832183838, "identity": 0.6772819757461548, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/8f9b554e4acf162d6e7f098c98bbc0303af6c5a710be169c14f7e98bff6553f1.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "904e41db7938132e1b8921963b0830a931c3c2647f3e408192b06757c7b13da9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6b1a5f9dd3ee4eeb578b81444f3b2f5955289ba0d2a4ccdcda4f66a7dd8ab791", "generation_seconds_batch": 23.280895702075213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7163355350494385, "target_z": 0.06835934814393818, "ce": 6.09409761428833, "genuineness": 4.519899368286133, "identity": 0.596338152885437, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/904e41db7938132e1b8921963b0830a931c3c2647f3e408192b06757c7b13da9.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "90899cdee1140e4fc48edd997be580cf79d0bbf8c5d0dc4c08c7fcd7a9cce6fb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ff50457afca00fb1c8eb84030a8036ea43fab48947314026b2035876a248a31f", "generation_seconds_batch": 23.464857763028704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6179900169372559, "target_z": -0.04980474568813394, "ce": 6.0305562019348145, "genuineness": 4.557555198669434, "identity": 0.6700166463851929, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/90899cdee1140e4fc48edd997be580cf79d0bbf8c5d0dc4c08c7fcd7a9cce6fb.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": "90b605c4fab42776fcc8bd56037462d2d2b46240ef1749dd8507b0ea84eaa8d6", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78666c429d82ce5c5ec73c387a924836b5546d562c01f1cce4071ea36d9e1313", "generation_seconds_batch": 16.690835547051392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 6.086154937744141, "genuineness": 4.356721878051758, "identity": 0.6749647855758667, "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_high/90b605c4fab42776fcc8bd56037462d2d2b46240ef1749dd8507b0ea84eaa8d6.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "90de33fc9eeb03a4b0f3d7fe22d82fa46f347da6527f13cee807e4468a91ef69", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2d550779ae1fdb31c1686960afb3c114038c136d223ac5145f7d54f913cc84f0", "generation_seconds_batch": 23.109901037998497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.046321392059326, "target_z": 0.4648439111363709, "ce": 6.403861999511719, "genuineness": 4.356721878051758, "identity": 0.5409637093544006, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/90de33fc9eeb03a4b0f3d7fe22d82fa46f347da6527f13cee807e4468a91ef69.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9103bddbbc147564d4a782f41c2eef32ddbd0d8806b268459a03aed0667a5f53", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1d68115faa1bb312f6a2ae22763381397d586a52a2301543547903a659c4b7c7", "generation_seconds_batch": 17.811416417942382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.793492794036865, "target_z": 3.765624928383835, "ce": 5.839932441711426, "genuineness": 3.785602569580078, "identity": 0.03648862987756729, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9103bddbbc147564d4a782f41c2eef32ddbd0d8806b268459a03aed0667a5f53.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "913a15480d24290c91f266b7680a90dc51a718e061458b949786c95849279c5e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a0bf4e005db90b0d87b4cbd293a8fa9283820fb98d569bf089d47df3a996350b", "generation_seconds_batch": 16.671934792073444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3291661739349365, "target_z": 0.8046875358080824, "ce": 6.165581703186035, "genuineness": 4.168440818786621, "identity": 0.7073541879653931, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/913a15480d24290c91f266b7680a90dc51a718e061458b949786c95849279c5e.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "91413fe446c04405fa495f9f2128f6214548c4a807ef6704f1ad9d139f2528c5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3de78096a3cfe72777cca3b159e6aa27e1e0c6ffc4ea31f84876ed191515f4f6", "generation_seconds_batch": 23.81462057295721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4150304794311523, "target_z": 2.109375071616165, "ce": 6.173524379730225, "genuineness": 4.13078498840332, "identity": 0.4498160481452942, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/91413fe446c04405fa495f9f2128f6214548c4a807ef6704f1ad9d139f2528c5.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "919286e1795eecdeef6ed7519fae60b1029cfcee5c708fdf59f45a081bfb913e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d155bbe08d14d8b9e7c4997a750b46a228a30dd1ba05313203127f375a2d4d35", "generation_seconds_batch": 22.68887325504329, "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": 5.95112943649292, "genuineness": 4.21237325668335, "identity": 0.5666964054107666, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/919286e1795eecdeef6ed7519fae60b1029cfcee5c708fdf59f45a081bfb913e.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "91b5ba771b9a2cc9f9a85840f91a7867bdd81804bcf1a16dd1652809c871a5aa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "54d4728d28c74ccdbf042142844a383fcb76a7d9e65e1a90b43b14a3e2272e62", "generation_seconds_batch": 22.725682756048627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.16794753074646, "target_z": 1.8125, "ce": 5.982900142669678, "genuineness": 4.356721878051758, "identity": 0.5236637592315674, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/91b5ba771b9a2cc9f9a85840f91a7867bdd81804bcf1a16dd1652809c871a5aa.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "91fac02c3196aac0d9d024c70810d4f1f8c1f878c4c551a56ffba6e6b806b700", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5daaddd75c227160bd889144c66ea32629878a86874c61369ea70af5ad352150", "generation_seconds_batch": 22.99340703105554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9598777294158936, "target_z": 1.5625, "ce": 6.054384231567383, "genuineness": 4.143336772918701, "identity": 0.169327974319458, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/91fac02c3196aac0d9d024c70810d4f1f8c1f878c4c551a56ffba6e6b806b700.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "932a2e6b5aecf5ff77d81cb580831af92d47e9fcaf68e96937a7c68c1be3bd1c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a4447f84fa8f6b287bd19e2b5bd5615ffb2cf16e6195d8f45acc1d831e97f1af", "generation_seconds_batch": 18.668961566057988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9829249382019043, "target_z": 0.38867181233585574, "ce": 6.221180438995361, "genuineness": 4.7081804275512695, "identity": 0.7266231775283813, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/932a2e6b5aecf5ff77d81cb580831af92d47e9fcaf68e96937a7c68c1be3bd1c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "9382ed5d12695c2af8255f6ba5acab36ee205cb8415da43cbb05d100a39afd04", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d0fa062a946646ce6aa44be321010ba611165bb525a2576ac92e99ed9d56adae", "generation_seconds_batch": 22.6588852860732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.10426139831543, "target_z": 2.9374997135353405, "ce": 5.927301406860352, "genuineness": 4.074300289154053, "identity": 0.5716854333877563, "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_high/9382ed5d12695c2af8255f6ba5acab36ee205cb8415da43cbb05d100a39afd04.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "93af6a8ffe233d6ff71748ac12389d115b5c280043d41f5e033170ffb1ab7feb", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1bd52ebc2e478775a459bf1fbb454309e18c3ffe3735034808017bddd6314634", "generation_seconds_batch": 17.258045624941587, "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": 6.308549880981445, "genuineness": 4.356721878051758, "identity": 0.32584208250045776, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/93af6a8ffe233d6ff71748ac12389d115b5c280043d41f5e033170ffb1ab7feb.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "93b519baa66d401ec432dc5ff55b3c1cca12dfff2d823a8f8272962c44b6a4b7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a53232069f0fa977a4d1efe73f8a551ac6b14559cfe6fc39082d9607bc75d979", "generation_seconds_batch": 21.594669436104596, "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": 6.149696350097656, "genuineness": 4.770940780639648, "identity": 0.6887710690498352, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/93b519baa66d401ec432dc5ff55b3c1cca12dfff2d823a8f8272962c44b6a4b7.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "942e68f011f6b45867019b05c5c9e5f4337faafbb75865c41448b11b06b67372", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fabdba78ebb47ac30a028a839458fb9870c67ea8c641915fe86b25702b2c0615", "generation_seconds_batch": 23.633638267987408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1828670501708984, "target_z": 0.6289062320959588, "ce": 6.284721851348877, "genuineness": 4.758388519287109, "identity": 0.692017674446106, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/942e68f011f6b45867019b05c5c9e5f4337faafbb75865c41448b11b06b67372.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "948938d9a999669a7047fcfb8b0709cb32568f3f6a0302a72e2526d079809373", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f6288d48d9f18dd58353d23c87c479b1cb7df2032d0cbb6abc600a64ba70f73a", "generation_seconds_batch": 23.048965709051117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7057695388793945, "target_z": 0.055664093832072124, "ce": 6.1020402908325195, "genuineness": 4.18099308013916, "identity": 0.3615497946739197, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/948938d9a999669a7047fcfb8b0709cb32568f3f6a0302a72e2526d079809373.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "948aafa875d2b6118395a7a974fe4bbae96cefca0f8cb93bffb4ba50abf5f125", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7d4eb54cdde79cffea1b305be2cad4bd6d5fa0cecb0a982c1febbd6b76a8b815", "generation_seconds_batch": 22.488092345069163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.442375183105469, "target_z": 3.3437501432323296, "ce": 5.8875885009765625, "genuineness": 4.055472373962402, "identity": 0.43724504113197327, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/948aafa875d2b6118395a7a974fe4bbae96cefca0f8cb93bffb4ba50abf5f125.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "95c80a43f5d03cedb4085c59dd087c5edee68eb2feb5943f0fb648c78bfe1573", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ff97a1c4e93aae4d7d0a8856e35cf808022d98c273f5210893e00f81a4b7f77b", "generation_seconds_batch": 25.37856071803253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.057410478591919, "target_z": 1.6796875358080825, "ce": 5.534139156341553, "genuineness": 4.356721878051758, "identity": 0.7891626358032227, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/95c80a43f5d03cedb4085c59dd087c5edee68eb2feb5943f0fb648c78bfe1573.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "962a00b46f72736c79bd9bfe7e65736fbba063f85fb8e4244205d48ddea10dfb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c74c67c827be288b29d7746f155f920b34c1e493fff332e01f21c7f6c5e993db", "generation_seconds_batch": 22.353658317937516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.897527694702148, "target_z": 3.890624928383835, "ce": 5.736677646636963, "genuineness": 3.660081624984741, "identity": 0.43260836601257324, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/962a00b46f72736c79bd9bfe7e65736fbba063f85fb8e4244205d48ddea10dfb.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "963e7f0eca0910f86fc432af9394d84927508a4c7a2f34b06b1f62a747aabb4e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b723cda88621cc8a2a01128bc54c1ba9ad7e6a6b2b118a3733958b52faa0ce2e", "generation_seconds_batch": 17.54144533595536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5112271308898926, "target_z": 1.0234373925757527, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.7385640144348145, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/963e7f0eca0910f86fc432af9394d84927508a4c7a2f34b06b1f62a747aabb4e.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "965a75faa20ce7d3fbee71bfc374052a2553e76a17f69523f63d931ee9d428f0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "641f5b39defd38ca384d96d365687cfcffcfee7fb6eec7fc9be29d79a56631e2", "generation_seconds_batch": 22.484702782938257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.010559320449829, "target_z": 0.42187507161616483, "ce": 5.542081832885742, "genuineness": 4.607763767242432, "identity": 0.6717779636383057, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/965a75faa20ce7d3fbee71bfc374052a2553e76a17f69523f63d931ee9d428f0.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "965da2ed6388dd13879c03fa24a2c045420d854d2d20a9bcda4d7bb8548cd274", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "23e13e1a76da493556ebdb52aa9e2b6ad809191cc7653123e51509b2e12719ff", "generation_seconds_batch": 23.024417896056548, "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": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.3409726023674011, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/965da2ed6388dd13879c03fa24a2c045420d854d2d20a9bcda4d7bb8548cd274.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": "9666fb88605b602a5704a31db2663972e9ccd8aaf84ce7659ff66ea37afe2859", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0610c2a8f92fdd160affa5c168c86ede49a13b1a3cb4bd5edafbdebe37b9a17c", "generation_seconds_batch": 17.428951447014697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.493056535720825, "target_z": 2.203124928383835, "ce": 5.8875885009765625, "genuineness": 4.18099308013916, "identity": 0.7313804626464844, "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_high/9666fb88605b602a5704a31db2663972e9ccd8aaf84ce7659ff66ea37afe2859.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9726dac6911bb923a5b7000b974d931a5579646fda8676f7955ea87299b9f5c2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f68ef546f21c895413690af210871cc1684b72526d53737c2948cf27ea408504", "generation_seconds_batch": 24.068337997887284, "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": 6.23706579208374, "genuineness": 4.808597087860107, "identity": 0.5673258304595947, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9726dac6911bb923a5b7000b974d931a5579646fda8676f7955ea87299b9f5c2.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "97c9cba70c9b30c0ab5546de9606c32b23e8a6fe98238e12ed4c7821378dbc9f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7256f90c3dbc4a92977dc2a03c859a32d5d3558cccae30755c5a3757a9a7d33e", "generation_seconds_batch": 16.37857794004958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.309659481048584, "target_z": 0.7812498567676703, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.7206699848175049, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/97c9cba70c9b30c0ab5546de9606c32b23e8a6fe98238e12ed4c7821378dbc9f.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "983d00ebe1d93319b5931d607469f4b246f43ce9483065d284ca1f9f5a72fd30", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7de8cefebda0622499394a07996ef13c9411fcd48436ad5b06f69952ca4684e1", "generation_seconds_batch": 17.389430526993237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049572467803955, "target_z": 0.4687501432323297, "ce": 6.221180438995361, "genuineness": 4.6203155517578125, "identity": 0.5065748691558838, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/983d00ebe1d93319b5931d607469f4b246f43ce9483065d284ca1f9f5a72fd30.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "98a3af7839914d1e96608e88db4c6c94af925a4b3629794b532cef726c5932ff", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1d95cc942de7423f8230f49dfa961661b055c29033eb566cf3915d0fd8e6f432", "generation_seconds_batch": 23.057995185954496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 5.967014789581299, "genuineness": 4.055472373962402, "identity": 0.751375675201416, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/98a3af7839914d1e96608e88db4c6c94af925a4b3629794b532cef726c5932ff.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "98a91d5e46ee4714099634f8c8871b815a3644f453270fab60f3424df197c061", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e6923630bd1285e083f2ef71dca0af41236c492ffcacd9ee205bb689eb7cac69", "generation_seconds_batch": 25.912419773056172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8428385257720947, "target_z": 1.4218750716161648, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.36542820930480957, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/98a91d5e46ee4714099634f8c8871b815a3644f453270fab60f3424df197c061.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "98e119286b5c18b752077af4b48f459e4adb1b33d412423c96627354214ed53b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "af3617024d0bbbf8c76b5fc93adb7a3f309786a4cebfbd44d2d302838ec17e0f", "generation_seconds_batch": 17.253113840008155, "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": 6.086154937744141, "genuineness": 4.720732688903809, "identity": 0.6014043688774109, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/98e119286b5c18b752077af4b48f459e4adb1b33d412423c96627354214ed53b.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "087e22ec868502cac821d7ec8dba21b8884b85325db231a155166c6bab1391ca", "generation_seconds_batch": 15.693938454031013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1145942211151123, "target_z": 0.5468750716161649, "ce": 6.4515180587768555, "genuineness": 4.570107460021973, "identity": 0.4960930347442627, "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_high/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "99c2a674152c5efe2aa57aae8dee781c71580bedf4545382b36621573c971069", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "308664cab622ce888ed7e0cc5c15934b87d9111fc1f6e0c7476329c6071ce46b", "generation_seconds_batch": 21.46896954299882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6607773303985596, "target_z": 1.2031249283838352, "ce": 6.157639026641846, "genuineness": 4.13078498840332, "identity": 0.7245928049087524, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/99c2a674152c5efe2aa57aae8dee781c71580bedf4545382b36621573c971069.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "99e342c1589d231e69a401f923dd40f118e197c289df6ed8568b15308f700a3e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5e4cdf1a0072b5043f5657e49ae24bb7f2d92702f2763fd056357369320fce42", "generation_seconds_batch": 23.298010800033808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532069683074951, "target_z": 2.25, "ce": 6.046441555023193, "genuineness": 3.735394239425659, "identity": 0.4421418309211731, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/99e342c1589d231e69a401f923dd40f118e197c289df6ed8568b15308f700a3e.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9a3f2b5e00efa0b41027363c89b5e02fc2cf5a2a716f54ecb39f58d5f99ac6c8", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "97aca75b349f09cb6b8c9e6a544cc6e9924bc525f3daf430419f02a47f372b1d", "generation_seconds_batch": 17.85402303491719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6353631019592285, "target_z": -0.02893063888494204, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.722530722618103, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9a3f2b5e00efa0b41027363c89b5e02fc2cf5a2a716f54ecb39f58d5f99ac6c8.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "9a665a67d914af893cb97fd423820b3d1e319fc2b19fb4e5ca63ce88665a8768", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e04c81676c44c15f0571e0fe537981a6a5e106fee1778751d8929743b8180abe", "generation_seconds_batch": 23.17735840496607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.935784101486206, "target_z": 0.3320311604797939, "ce": 5.935244083404541, "genuineness": 4.532451152801514, "identity": 0.5888117551803589, "hypothesis": "La porte était encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9a665a67d914af893cb97fd423820b3d1e319fc2b19fb4e5ca63ce88665a8768.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "9a6aeb22549b1b694c9357d76a3a9ebff702358d4c394113d9373c88784ea12e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1bf81cecd0551fe23d03b1fa17820191dec0013c5ad8e8c630429dd4d0561d7e", "generation_seconds_batch": 17.253113840008155, "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": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.5328516960144043, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9a6aeb22549b1b694c9357d76a3a9ebff702358d4c394113d9373c88784ea12e.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "9a808c60c332639674c38c1bd283118018aabc92e726da2751a576a3034215ee", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ff99c07b2b5e1bafa00501c56a2619efbf9b1779d84304e644ec97e28e69e0fc", "generation_seconds_batch": 15.75975651293993, "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": 6.252951145172119, "genuineness": 4.325342178344727, "identity": 0.39565664529800415, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9a808c60c332639674c38c1bd283118018aabc92e726da2751a576a3034215ee.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "9ac2df839eba59b51bc0199d2565c64a150a1707761ca2dc02c9bc990142754b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "30ac8ea6e9864a35aeba780010e8127e13304df738670d8ceb21233d96cfd8a1", "generation_seconds_batch": 17.98971182096284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6246954202651978, "target_z": -0.04174807037405409, "ce": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.6533540487289429, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9ac2df839eba59b51bc0199d2565c64a150a1707761ca2dc02c9bc990142754b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9ae7dab379cca6936e3d950c8ed0ce1efb62a5c002d6999b9650e8f2bed7b8d9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9b7a955b08425454c3b28c69492a3ecd50347906a1e2187bf84d6df7b79b0de1", "generation_seconds_batch": 22.296903026057407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.870182991027832, "target_z": 2.6562498567676704, "ce": 6.022613525390625, "genuineness": 4.005263805389404, "identity": 0.5540857315063477, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9ae7dab379cca6936e3d950c8ed0ce1efb62a5c002d6999b9650e8f2bed7b8d9.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9b7be4bd78036120d2e1b168db2f2fcfa60431614cd9710b3a891cb4bb5560d9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b3bf8cd39bad589851cdd4bebf9d3dd1a159e30a28f432d7eabc7b6b900f8ae6", "generation_seconds_batch": 22.296903026057407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.98722243309021, "target_z": 2.796875071616165, "ce": 5.696964263916016, "genuineness": 3.7793264389038086, "identity": 0.5462995767593384, "hypothesis": "De deur 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_high/9b7be4bd78036120d2e1b168db2f2fcfa60431614cd9710b3a891cb4bb5560d9.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "dc084c724969a0ef067b6b9aee72bcb464534ef381b37ec1e4d761b81fd2d7fb", "generation_seconds_batch": 17.855580860050395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680284023284912, "target_z": 1.2265626074242473, "ce": 6.284721851348877, "genuineness": 4.306513786315918, "identity": 0.7187361717224121, "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_high/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9c25b4f30f89636f85d1d6b5cc4c2c98e22112010c78b77ce3dcad429bdd3833", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bd8d3c6778e6c36b4d5e8c6ba0a6601e4b404c4d171428140f1052df1320249f", "generation_seconds_batch": 17.639247552026063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.496886968612671, "target_z": -0.19531246419191758, "ce": 6.1894097328186035, "genuineness": 4.457139015197754, "identity": 0.7432565689086914, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9c25b4f30f89636f85d1d6b5cc4c2c98e22112010c78b77ce3dcad429bdd3833.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "9c3bcd460a181345652af49aa29be6c48f6fcc2785cd4f12a461b90b62a065d8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "314d733a29cfc4d072651d1273ef7d009d52afc6e9963aa07163a9838cbdd2a1", "generation_seconds_batch": 25.081651952001266, "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": 6.292664527893066, "genuineness": 4.65797233581543, "identity": 0.5843799114227295, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9c3bcd460a181345652af49aa29be6c48f6fcc2785cd4f12a461b90b62a065d8.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "9c9295cbc3211fa92dd7ffa6b18f20f640054d94d39439e97837df3f6983851e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "275f801d5820f5b3d4ff5d55cc671e4307a198afd033293399ebef63b13d5bc6", "generation_seconds_batch": 23.025358312996104, "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": 6.252951145172119, "genuineness": 4.6203155517578125, "identity": 0.652795672416687, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9c9295cbc3211fa92dd7ffa6b18f20f640054d94d39439e97837df3f6983851e.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "6700e759db0a4f3f69ff4738deb52abda7ebbaa790249c059349cd89d9b25b7f", "generation_seconds_batch": 16.857097771018744, "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": 6.24500846862793, "genuineness": 4.218649387359619, "identity": 0.4221475124359131, "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_high/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9c97f79fa715a010b85ca797a9374a45ad1b6d11e6fe2385be3e19a1004e377c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "da4e26346bd1ace20a8e53453fe7e487410b525c6a8559bedeba7bc0e2e0ec28", "generation_seconds_batch": 23.82093454990536, "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": 6.165581703186035, "genuineness": 4.356721878051758, "identity": 0.7318946123123169, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9c97f79fa715a010b85ca797a9374a45ad1b6d11e6fe2385be3e19a1004e377c.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": "9ccfee94abff6425fd4d38bf12ef72a0c41f0133f21c70f251e9fd0d37001df5", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d8b4318fcaf53171ee8c939832bdd96b94f175b0d493d23211b765a0706e84d", "generation_seconds_batch": 17.428951447014697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.688153862953186, "hypothesis": "Mon message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9ccfee94abff6425fd4d38bf12ef72a0c41f0133f21c70f251e9fd0d37001df5.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9cf112d3ddd2259aceaae68c9e95140b25f5aabff5f5e68a369a4dd409f2a4e9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "413a3a2141ff6d2598779b46ca13bc761560d4a392473d98eb7aeab063e9c4ea", "generation_seconds_batch": 24.581176512059756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.316161632537842, "target_z": 0.7890623209595878, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.6557068824768066, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9cf112d3ddd2259aceaae68c9e95140b25f5aabff5f5e68a369a4dd409f2a4e9.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9d2eb23fe46a53fb2af9e730a526a9684a9cf6fb738f6aa87e7c148be1b85b66", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0db2e8d096158c5ecac710a9e6126706d65ab99f0a7d0d71ba3e47ecf36adeef", "generation_seconds_batch": 17.752706032944843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.719296932220459, "target_z": 1.2734373925757527, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.4946049153804779, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9d2eb23fe46a53fb2af9e730a526a9684a9cf6fb738f6aa87e7c148be1b85b66.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9d645f388378630ca092ef9abde7032c0bea735792b234d93d6ddc855cb7c5cc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "85d4418a4d83efb80dbe812ae24864982564acd51d53bcfc1361fc547bfc60ef", "generation_seconds_batch": 23.82584017305635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0769169330596924, "target_z": 1.7031249283838352, "ce": 6.086154937744141, "genuineness": 4.13078498840332, "identity": 0.4720965027809143, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9d645f388378630ca092ef9abde7032c0bea735792b234d93d6ddc855cb7c5cc.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "9dbb1d5f10388b78f2744988499b8e0d17da9500188ecbee510d0cb90946b22c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "690c13c8931113cf8364bf493bea9cde4597024016d04e328c7b45e4ce9795fe", "generation_seconds_batch": 24.29668874805793, "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": 6.078212261199951, "genuineness": 4.432034492492676, "identity": 0.7028339505195618, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9dbb1d5f10388b78f2744988499b8e0d17da9500188ecbee510d0cb90946b22c.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "9e22432a8f0b95862ce9d29a784f46120d6f79d4847a20974a0c286b4be3a3fa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "49d41fe06b086e4bc250c00e1a3d01b1fab54f2d5a8474df48360f30584ee90b", "generation_seconds_batch": 24.37908913800493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.946873426437378, "target_z": 1.5468750716161648, "ce": 5.0178656578063965, "genuineness": 4.21237325668335, "identity": 0.4423779547214508, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9e22432a8f0b95862ce9d29a784f46120d6f79d4847a20974a0c286b4be3a3fa.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "9e2e6f71f1b78195c79ebfeef1b89a2cab50db04023447bd01092d872fc0afb4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "9c6c3c39f79548e412f4ed4a2a25572ee10299cd122ed3f09c1b19b1115afe7a", "generation_seconds_batch": 21.68409943394363, "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": 5.689021587371826, "genuineness": 4.432034492492676, "identity": 0.7291116714477539, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9e2e6f71f1b78195c79ebfeef1b89a2cab50db04023447bd01092d872fc0afb4.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "9f1d7df07f0a1c35ebecb6db562573276a34dab3270ed32ff0c6781d0fd1f1bb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c365907690886868f3758fdd31c15843a18102392a0aa44e3de1fc83ee783df1", "generation_seconds_batch": 23.19902868999634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8558428287506104, "target_z": 1.4375, "ce": 5.5539960861206055, "genuineness": 4.293961524963379, "identity": 0.7621597051620483, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9f1d7df07f0a1c35ebecb6db562573276a34dab3270ed32ff0c6781d0fd1f1bb.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "9f69b224562739919bb17043ad8ec6ee272087eb6c3dccbdbd7841481e8aea8e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "125344cc75a9b45a09aa3de5c66341d756f44100a3a36cfe1ea0125931e03088", "generation_seconds_batch": 22.159214230021462, "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": 6.213237762451172, "genuineness": 3.8295347690582275, "identity": 0.39091384410858154, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9f69b224562739919bb17043ad8ec6ee272087eb6c3dccbdbd7841481e8aea8e.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "9fa130779e45b548ca7ffc015050f23d0a596590611e30e60e0fa59283858b31", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e138f4c5b11d49a04871b0e08cd55118419fbd54da1b56b19fe89842f77611f8", "generation_seconds_batch": 22.50709457800258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1094279289245605, "target_z": 1.7421875358080825, "ce": 6.078212261199951, "genuineness": 4.218649387359619, "identity": 0.4632992744445801, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9fa130779e45b548ca7ffc015050f23d0a596590611e30e60e0fa59283858b31.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "9fb6cdb95d8569397a0852b6760b4cba8a4bfb71c20653ff5d024f73c16161cf", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9252318a660f644b986c1ea7257ea64fcbc5b843ed3054dbc1c4ae3fbd8aca0d", "generation_seconds_batch": 16.987020627013408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7789191007614136, "target_z": 0.14355474568813395, "ce": 6.157639026641846, "genuineness": 4.8588056564331055, "identity": 0.6386026740074158, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/9fb6cdb95d8569397a0852b6760b4cba8a4bfb71c20653ff5d024f73c16161cf.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "a03865f4efe54d2f9dcd3cf104856af494e63f966619c7b7c96d33ba1499fad3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "92fd32512733f9e9f85cf167ccc3743750d32e0fed08f17423f16b5ff18b9b78", "generation_seconds_batch": 26.706156763015315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.821183204650879, "target_z": 0.19433590616792787, "ce": 6.292664527893066, "genuineness": 4.256305694580078, "identity": 0.5659136176109314, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a03865f4efe54d2f9dcd3cf104856af494e63f966619c7b7c96d33ba1499fad3.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": "a0fb4e7ccfa057e86111a9b8a94dc5f9ece4707866b2ccd4d941723cd36d7b12", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "07b120f243ccad378ba0f337d1b4d2a1b1e3df2216d1f017b01dce6a0b190d33", "generation_seconds_batch": 16.434750373940915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.206960678100586, "target_z": 1.8593750716161648, "ce": 6.229123115539551, "genuineness": 4.325342178344727, "identity": 0.6338584423065186, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a0fb4e7ccfa057e86111a9b8a94dc5f9ece4707866b2ccd4d941723cd36d7b12.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a1333dbda2bac53750618c273ed4725b303ad56c4ee8aa763cd1947f6b41d814", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c295fd90979f211b013b4d93d1a2e8e21cc2232a465219c9506841f6ed9fbf69", "generation_seconds_batch": 23.81462057295721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50606107711792, "target_z": 2.2187501432323296, "ce": 6.054384231567383, "genuineness": 4.030367851257324, "identity": 0.39277511835098267, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a1333dbda2bac53750618c273ed4725b303ad56c4ee8aa763cd1947f6b41d814.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "a1639aa3704aa96886d5b25035603b31f46100a32a36dae21a267965e4ab9bf6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4f609542532739263d7323c6a040a4e3647e3e84607e97c23d45063670d04990", "generation_seconds_batch": 17.35070348205045, "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": 6.109982967376709, "genuineness": 4.758388519287109, "identity": 0.6641619801521301, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a1639aa3704aa96886d5b25035603b31f46100a32a36dae21a267965e4ab9bf6.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "a19e83ccaa4da4ac5149c9bbffc546d69060e4a0d8e7ce08bb7b3e00032946ac", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1f0169a7c6b951537ec2519f6722f4e50b0de242dc3f8966d72d2199a950eae2", "generation_seconds_batch": 24.32273885491304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9048988819122314, "target_z": 0.29492195556818546, "ce": 6.308549880981445, "genuineness": 4.457139015197754, "identity": 0.47222915291786194, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a19e83ccaa4da4ac5149c9bbffc546d69060e4a0d8e7ce08bb7b3e00032946ac.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a1f2dd853cbfe199eaa88c3a48a5302fd64f9828dbb0edd0013f5645c6b8e622", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "86e2f9e8555e9b1cf583c6dc534ba0b94c07bf7003d8beebeff4a96eee355f69", "generation_seconds_batch": 17.793326555052772, "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": 4.413206577301025, "identity": 0.3362773656845093, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a1f2dd853cbfe199eaa88c3a48a5302fd64f9828dbb0edd0013f5645c6b8e622.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "a21e3c2dce431c7ec6d6479856bd96885e6adc06bcbbb58fea76df1d08c3f647", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1197675b5a6b750170d363a0f1abcd419e40283d5dc15ba616c4606b0e147fe3", "generation_seconds_batch": 24.012570268008858, "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": 6.078212261199951, "genuineness": 4.595211505889893, "identity": 0.6276260614395142, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a21e3c2dce431c7ec6d6479856bd96885e6adc06bcbbb58fea76df1d08c3f647.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "a26f6f5bbeb5677265d3ae21ec213a703f5501c9322806249ab3f1baaf0a6318", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "de9a3ac95a3a5d76f175a57ffc976e72b86caba9cb152954cdb5ff3eaa997c8f", "generation_seconds_batch": 22.945971267065033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1145942211151123, "target_z": 0.5468750716161649, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.7214743494987488, "hypothesis": "Die Tür 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_high/a26f6f5bbeb5677265d3ae21ec213a703f5501c9322806249ab3f1baaf0a6318.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "a28202c2f98d58dda395af9927c82109e817aa562e530dd04152f061cf45eeca", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8c47084e7a2e9b2c27091b5481c5de5c59d17b4bda4566f13972b4e0cca08acd", "generation_seconds_batch": 17.234165358007886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 6.23706579208374, "genuineness": 4.268857479095459, "identity": 0.6668511629104614, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a28202c2f98d58dda395af9927c82109e817aa562e530dd04152f061cf45eeca.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a2b1f5201b72148e89726d1d6f2e7a13941ef50ba5515cf4c73ffd811e0f54da", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3f59db25658412091f998f3a739a1566ff9906cbc31686419c8fda0384de5aac", "generation_seconds_batch": 18.1122174190823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370041847229004, "target_z": 2.015624928383835, "ce": 6.109982967376709, "genuineness": 4.193545341491699, "identity": 0.23909515142440796, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a2b1f5201b72148e89726d1d6f2e7a13941ef50ba5515cf4c73ffd811e0f54da.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "a30f2b96bf8b87d896e0e9e207bdab1a523a2dfa10e9c3ff2df1ef88deab5f5a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ef48c46ecab69108603c81afcfed8cc093ce9d71370f2e256fde598ddafe6335", "generation_seconds_batch": 23.068849861971103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.862344980239868, "target_z": 1.4453124641919175, "ce": 6.0146708488464355, "genuineness": 4.469690799713135, "identity": 0.7262083292007446, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a30f2b96bf8b87d896e0e9e207bdab1a523a2dfa10e9c3ff2df1ef88deab5f5a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "a34b3983948d4880e5ed37536f33af040b5ababf7f5570acb1632555adce7fa5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f90b2deb1d32789bd458d687411dc6cf26ee125f48e9b389bbb2109b9ca7e1e0", "generation_seconds_batch": 22.84008495497983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.836336135864258, "target_z": 1.414062320959588, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.5536777973175049, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a34b3983948d4880e5ed37536f33af040b5ababf7f5570acb1632555adce7fa5.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "a373f2d13b34eb57ea33e9fcbb027cee2d1d1f8f634a3302ba3e4a50d79f4fb0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3e9a8840d13ad3464702af3cab9c696e6456555c6be4142592a3cc8c5d3a4777", "generation_seconds_batch": 23.25120004697237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.797323226928711, "target_z": 1.3671875358080825, "ce": 6.149696350097656, "genuineness": 4.2751336097717285, "identity": 0.504416286945343, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a373f2d13b34eb57ea33e9fcbb027cee2d1d1f8f634a3302ba3e4a50d79f4fb0.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": "a392fa6c80c23b90d0d961b22d3f76300801e6ca536bbb7fd2fdf38fc44bd64e", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8291e74a91a4e2ff413b1c3fc44fb9ea8ab81162eab9cb0a1cf9fe832e43c878", "generation_seconds_batch": 22.667473532957956, "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": 6.197352409362793, "genuineness": 4.507347106933594, "identity": 0.5468854308128357, "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_high/a392fa6c80c23b90d0d961b22d3f76300801e6ca536bbb7fd2fdf38fc44bd64e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "a425e47c85e68c1505e6a5271f539b2edad435481afadc8d20178b196a68fbe9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5e0ce7906c9b7002e8f1009dda919981349686ec0f52d1da6fc0725438f38fbc", "generation_seconds_batch": 23.024960463983007, "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": 6.1894097328186035, "genuineness": 4.256305694580078, "identity": 0.5201880931854248, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a425e47c85e68c1505e6a5271f539b2edad435481afadc8d20178b196a68fbe9.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": "a463b1888cf48dbc2a4a3a5fe5ae0355f8dc320de2773edee968f168c2ff590f", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "336d12f78bd901900963440566e2cd04fcb55ded4518c212f76beebb1e71a63f", "generation_seconds_batch": 17.14151410292834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6542751789093018, "target_z": 1.1953124641919175, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.7101892828941345, "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_high/a463b1888cf48dbc2a4a3a5fe5ae0355f8dc320de2773edee968f168c2ff590f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "a494ab82d19a9481e8a518e09eb04fc22f82d12d63a3e2985d0d827488efb4ec", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8fa3a8255dda26b693cecbe16f0c0139f12c3df8c6ba7c9ce699f73450be5495", "generation_seconds_batch": 23.388810685952194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 5.764476776123047, "genuineness": 4.168440818786621, "identity": 0.6024734377861023, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a494ab82d19a9481e8a518e09eb04fc22f82d12d63a3e2985d0d827488efb4ec.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": "a4c963f4980a205359a6f62d8b44efc02acdb70eff273e3730f8167304e3140e", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ff7585f233731480bf51820979da22e50fe24e6fb05e788051ed539590131bad", "generation_seconds_batch": 17.993950573028997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.089921474456787, "target_z": 1.7187501432323298, "ce": 6.403861999511719, "genuineness": 3.9425034523010254, "identity": 0.5091218948364258, "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": 0.120000000000001, "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_high/a4c963f4980a205359a6f62d8b44efc02acdb70eff273e3730f8167304e3140e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "a4e6e0124a10d8ed7e83592808577e16d267ddc6b9cfe34d125f8a0d78afde9e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9640ae26356d77688b144cf71087da1b437313057781a153466aa5253a4225fa", "generation_seconds_batch": 24.32273885491304, "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": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.3632860481739044, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a4e6e0124a10d8ed7e83592808577e16d267ddc6b9cfe34d125f8a0d78afde9e.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "a520f136591f3d774485a7c669ec0b76dc14c4c899b9aae8fd4add4989174bf5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "308eade90b1c4b7b2be45df2d98450a187acf11dccd0e44a557a543cd5f7c0a2", "generation_seconds_batch": 22.89550835103728, "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": 6.23706579208374, "genuineness": 4.607763767242432, "identity": 0.5651921629905701, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a520f136591f3d774485a7c669ec0b76dc14c4c899b9aae8fd4add4989174bf5.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "a53cc9b0515185ef5a570d5aeff81cbc0fd42aa8e7b866a6ad77876aeba3a404", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c947a442f4eddec3fae27338242e7afc5d9317edc33004ada9d6b96d23f5aa2b", "generation_seconds_batch": 24.37908913800493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7713146209716797, "target_z": 1.335937679040412, "ce": 5.839932441711426, "genuineness": 4.582659721374512, "identity": 0.7176234722137451, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a53cc9b0515185ef5a570d5aeff81cbc0fd42aa8e7b866a6ad77876aeba3a404.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a58de62c3583084e049fdcdff37dd23014b61e5ebec437ff075058a8b0757769", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3288351c65c9bac63ca40522d561759144ff0f6d7a4b363545699987e794c607", "generation_seconds_batch": 18.013003390049562, "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": 6.324435234069824, "genuineness": 4.218649387359619, "identity": 0.26305076479911804, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a58de62c3583084e049fdcdff37dd23014b61e5ebec437ff075058a8b0757769.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "a58f9a9f42d7cab1f8454bb12dba028b47206a1396e85a2cc90859607714e790", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9cfc5f237fb4f31e1c5ac1ae1e8f47979e23e09c5b60904021e7d16a8a50493d", "generation_seconds_batch": 18.523241147049703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1828670501708984, "target_z": 0.6289062320959588, "ce": 5.847874641418457, "genuineness": 4.532451152801514, "identity": 0.7850067615509033, "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_high/a58f9a9f42d7cab1f8454bb12dba028b47206a1396e85a2cc90859607714e790.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a6b7832ec00ae8755718904681467f8ad0f0fb7787f634381f7d0824859559ca", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "34f5394efa7b48964acada058dc1ce11e52c03d2501277b274bd60695dced6eb", "generation_seconds_batch": 21.794991574948654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.805161237716675, "target_z": 2.578124928383835, "ce": 5.95112943649292, "genuineness": 4.18099308013916, "identity": 0.4654696583747864, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a6b7832ec00ae8755718904681467f8ad0f0fb7787f634381f7d0824859559ca.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "a7c2a1bf63ca7bd3dfa2ad4b498f91a858b83322ed927e17fc548134e08ef340", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "483dff6e2404bd32334f1f5ab8b0253f98e66d9036c21d2c3b46d682868f22b1", "generation_seconds_batch": 16.698649651021697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.350008487701416, "target_z": 2.0312498567676704, "ce": 6.117925643920898, "genuineness": 4.174716949462891, "identity": 0.8258259296417236, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a7c2a1bf63ca7bd3dfa2ad4b498f91a858b83322ed927e17fc548134e08ef340.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": "a826a6da47ff3c8991bd47a1af3df029c45d3cc00c113c2a71a0dec0e318efe3", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c5625df5d86e5d7fecfa0965b1c5b7da279ab0fccf4829ad4c3a48ca3f82b0a", "generation_seconds_batch": 16.665557779022492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.011895179748535, "target_z": 1.625, "ce": 6.2767791748046875, "genuineness": 4.105680465698242, "identity": 0.4108601212501526, "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_high/a826a6da47ff3c8991bd47a1af3df029c45d3cc00c113c2a71a0dec0e318efe3.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "a84c6b117c4df873055582b9d5e37b436bf73d25ec1801351c6acc217187e2cb", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5c6fbe12eee0759b146746addf4772ae6c602043780593c4ed6d4a97fb2fa120", "generation_seconds_batch": 16.751722809043713, "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": 6.23706579208374, "genuineness": 4.011539936065674, "identity": 0.36047953367233276, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a84c6b117c4df873055582b9d5e37b436bf73d25ec1801351c6acc217187e2cb.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "a89599cc67c57e8696bd12b1656d52609ab5defd90be695f14b06a67abfab609", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "01bd647733d46cca4be7b6c8e3f63c92da8c7aabe59dbd19e3254804f6769c60", "generation_seconds_batch": 17.084808100946248, "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": 6.125868320465088, "genuineness": 4.833701133728027, "identity": 0.5928490161895752, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a89599cc67c57e8696bd12b1656d52609ab5defd90be695f14b06a67abfab609.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "4a8bb83733cfb4479e96941cd880a2783932f96fe94864376e9f1a53909c1829", "generation_seconds_batch": 16.53020427795127, "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": 6.197352409362793, "genuineness": 4.055472373962402, "identity": 0.7328871488571167, "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_high/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "a8edf1f5524647f269ee7f7121e3bf40693227b7817c7697aac4e4fbc034c4fb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f25dda9c2886a50f559115e62dab1ba5de95c8d68b4784c4ea4a0701e6fc8ae3", "generation_seconds_batch": 21.598848772002384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.610095977783203, "target_z": 2.3437501432323296, "ce": 5.927301406860352, "genuineness": 4.118232727050781, "identity": 0.656730055809021, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a8edf1f5524647f269ee7f7121e3bf40693227b7817c7697aac4e4fbc034c4fb.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "a90d6ed9b00269193dfe44dc7bb11aebf62d3607e5ea3cfc8220902c84076586", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2bf3b025f2dc350107c3553b82178799a17bd0c8c2c06541e1cccc9886cd9c4f", "generation_seconds_batch": 23.63948260503821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5190653800964355, "target_z": 2.234375071616165, "ce": 6.213237762451172, "genuineness": 4.118232727050781, "identity": 0.6789058446884155, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a90d6ed9b00269193dfe44dc7bb11aebf62d3607e5ea3cfc8220902c84076586.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "060f5fc7685fed7dde0f9981818c3a5fcd9342d7304173b699773a458189e543", "generation_seconds_batch": 17.169369822950102, "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": 6.141753673553467, "genuineness": 4.381826400756836, "identity": 0.47769129276275635, "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_high/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "a961d38a520d10051e2dd6806acc3d402839c73acd8a98d98e4a32a8eab7886b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d6fd95333e4f3b430a6afd8012a99d5e876a304df563f582aadcd3839d39de1c", "generation_seconds_batch": 16.96842769498471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.088585376739502, "target_z": 0.5156249283838351, "ce": 6.0305562019348145, "genuineness": 4.532451152801514, "identity": 0.5380423069000244, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a961d38a520d10051e2dd6806acc3d402839c73acd8a98d98e4a32a8eab7886b.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "a99f478a42c5ef9a93e97539fc398de4c9fa270387d59be4cda7003635ba345d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f3c314bb699d4cfb861abce41b4a24e2728ccd00b8fe582b1701fe930e431faf", "generation_seconds_batch": 17.08283402491361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.641270875930786, "target_z": 1.1796875358080825, "ce": 6.133810997009277, "genuineness": 4.3127899169921875, "identity": 0.8117479085922241, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a99f478a42c5ef9a93e97539fc398de4c9fa270387d59be4cda7003635ba345d.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "a9b0d2fc59153941ac7a3a25ec4d21a9d1cd0f5070dee37137a347b5db9b9eba", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f26752c0f939005df41681df7075a2032ba4e46b77ce73362d5d747a170a0db7", "generation_seconds_batch": 23.33582736097742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 6.0146708488464355, "genuineness": 4.218649387359619, "identity": 0.6224864721298218, "hypothesis": "De de 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_high/a9b0d2fc59153941ac7a3a25ec4d21a9d1cd0f5070dee37137a347b5db9b9eba.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "a9f332c0bdfe548101236d8e29acb8a37117d88b8f684038079ed002fc555228", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a98968565feae72fc40a5a7cc86a17200edfe84a8e3aec4f2ee63004edb97af1", "generation_seconds_batch": 16.797234767931513, "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": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.4963197112083435, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/a9f332c0bdfe548101236d8e29acb8a37117d88b8f684038079ed002fc555228.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "aa53b6e4ddd2f7bc065040592806c790b34c951136ad640ecbc21f65b1e60a09", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "dc848f0e730bd52b6d5aebbbf42d0bd6071c5ea382f4804aeaed6ffa5dd72779", "generation_seconds_batch": 16.595890511991456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.317497730255127, "target_z": 1.9921875358080825, "ce": 5.808161735534668, "genuineness": 4.149612903594971, "identity": 0.47737860679626465, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/aa53b6e4ddd2f7bc065040592806c790b34c951136ad640ecbc21f65b1e60a09.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "aaee971b9ce482418e9e50f368edc66245f4cf2ab9cfaa274ebed4e0c9989df4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4db70e98276475e872c608f42b6fc1383b9534d40c611a908823c0e778ec6519", "generation_seconds_batch": 23.51663255598396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.039239883422852, "target_z": 2.859375071616165, "ce": 5.02977991104126, "genuineness": 3.9927117824554443, "identity": 0.7528061866760254, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/aaee971b9ce482418e9e50f368edc66245f4cf2ab9cfaa274ebed4e0c9989df4.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "aaf06fff3bd14c1ee459aac3cd88f7a97ff2d03e76259560fd8fdb7747c5bf39", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ac9cadf7ebaa7e92d1adb69b44b2ac04af49dbc6b118a3d22954458ab595bce5", "generation_seconds_batch": 16.96842769498471, "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": 6.1020402908325195, "genuineness": 4.532451152801514, "identity": 0.5727782249450684, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/aaf06fff3bd14c1ee459aac3cd88f7a97ff2d03e76259560fd8fdb7747c5bf39.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "aaf47102b2b3f93b3d60bfe8deb6397fccd2db966e19acafbeaea79ac9f782b7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "be650e0c061d34a4eab84e31bb9aaaaff4dedd9290e17bd8096d647aa33a7c7c", "generation_seconds_batch": 16.290564085938968, "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": 6.181467056274414, "genuineness": 4.6203155517578125, "identity": 0.5979119539260864, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/aaf47102b2b3f93b3d60bfe8deb6397fccd2db966e19acafbeaea79ac9f782b7.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ab3b068c9aa75f1bf981d1b65708cb7bff47f5ee3ffa7dead0594bd15d8f52b7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "29a48d6c3555a8691d54c90cf899fe6fdeba87aa2cf30d2de5680bee59f526f9", "generation_seconds_batch": 22.229893729905598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5984272956848145, "target_z": 3.5312498567676704, "ce": 5.855817794799805, "genuineness": 3.7918784618377686, "identity": 0.44266143441200256, "hypothesis": "De deur 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_high/ab3b068c9aa75f1bf981d1b65708cb7bff47f5ee3ffa7dead0594bd15d8f52b7.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "ab60695edaa09277adc78b22eed5cc7f32c5ba1e784e33ded7e8a9def14c8ad8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bb4b508c14f9e9e25a7ebbc4211c983e46891827a9cff7bbb7175bce0ee0c5c3", "generation_seconds_batch": 16.489052001968957, "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": 6.006728172302246, "genuineness": 4.331617832183838, "identity": 0.6992233991622925, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ab60695edaa09277adc78b22eed5cc7f32c5ba1e784e33ded7e8a9def14c8ad8.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ab695480329ddff9ee06e7a541ba2c1afb24a160c56aef3417d84a35f26a9643", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e03ae140bb7c8e0bdea334432b764f0f484a01f5db377857b9616416941e5883", "generation_seconds_batch": 24.48482840601355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6282663345336914, "target_z": 1.164062320959588, "ce": 6.062326908111572, "genuineness": 4.281409740447998, "identity": 0.6922810673713684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ab695480329ddff9ee06e7a541ba2c1afb24a160c56aef3417d84a35f26a9643.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "bf268292fce04d694be475d5e2fc3f40545ea7b99eac5cb3b884995d4b21cc98", "generation_seconds_batch": 18.48441807902418, "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": 6.3641486167907715, "genuineness": 4.118232727050781, "identity": 0.45582374930381775, "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_high/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "abd77fc2387016f9540dc873f481497e3ef69850e18f240cfab5463c06db9b20", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "bf5407fe8801e9919bdf9fe234e780e5ca728825f7d7d57ecb130651cfc4b147", "generation_seconds_batch": 23.51663255598396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.130270481109619, "target_z": 2.9687501432323296, "ce": 5.665193557739258, "genuineness": 3.9425034523010254, "identity": 0.5254306197166443, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/abd77fc2387016f9540dc873f481497e3ef69850e18f240cfab5463c06db9b20.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": "ac60c10fbe327e702638e08d4c08591b84ccc9aa93c72aedad5e8540176613fc", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d79f5fd731f859c242c110014ebb7e80cd27ea46b7608240ef1648b108123cd", "generation_seconds_batch": 23.88780442602001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.089921474456787, "target_z": 1.7187501432323298, "ce": 5.665193557739258, "genuineness": 4.118232727050781, "identity": 0.4471394121646881, "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_high/ac60c10fbe327e702638e08d4c08591b84ccc9aa93c72aedad5e8540176613fc.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "ad35fb85000ff77d3dabbc20768f9c8eb0c4694ba50a6a84077023189e9e685e", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "de23eeb60381489ad7c936927ed4c61006c1769099c0dcf5f06acfaf6e68791c", "generation_seconds_batch": 18.09574145905208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.689920425415039, "target_z": 0.03662106913194333, "ce": 6.499174118041992, "genuineness": 4.250029563903809, "identity": 0.47497397661209106, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ad35fb85000ff77d3dabbc20768f9c8eb0c4694ba50a6a84077023189e9e685e.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ad45faaf73c9b563b12756608c5f75fcc02e3491c62548dfc5b3ead942f466c0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3615aa71bc67fffdde01c2260e24cf3503c5701b364f029a15a562586a0c30ef", "generation_seconds_batch": 17.705933527089655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.232969284057617, "target_z": 1.8906249283838352, "ce": 6.133810997009277, "genuineness": 4.111956596374512, "identity": 0.7366001009941101, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ad45faaf73c9b563b12756608c5f75fcc02e3491c62548dfc5b3ead942f466c0.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "ad4ca7d92489b6e4580b565958322dc24c1b1005f82720176be1b92311b7b08b", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3e01801fcf8d80617ff8ed4f62421ff26d40aa2ad546910cb3e10d4b688d7363", "generation_seconds_batch": 17.573196347919293, "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": 6.23706579208374, "genuineness": 4.507347106933594, "identity": 0.5264068245887756, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ad4ca7d92489b6e4580b565958322dc24c1b1005f82720176be1b92311b7b08b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "ad5f85bd791d7b1e6cbbe521ca76ccf8755bee00f89af490a2588a4f21837468", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "34e188fadfcd5b73be62ad561779ed6712c495d822a18f95e4d47642e1526b36", "generation_seconds_batch": 17.404954246012494, "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": 6.038498878479004, "genuineness": 4.545003414154053, "identity": 0.7218598127365112, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ad5f85bd791d7b1e6cbbe521ca76ccf8755bee00f89af490a2588a4f21837468.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "ad97f44fc4a360f5881c6d3aa739e1914ba564760e17b02e0651ccc36362f312", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2829910d5fc6570ccca41fdc1dafeefb15cda2e66a75c015b137ba76ef697dad", "generation_seconds_batch": 18.25526084995363, "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": 5.974957466125488, "genuineness": 4.281409740447998, "identity": 0.49310413002967834, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ad97f44fc4a360f5881c6d3aa739e1914ba564760e17b02e0651ccc36362f312.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "ada5638b6650fe5b4db6da63782ba0a415ca7f0da9400c41a04f5457c981d4ce", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6740cae7b05b074c1433bb46aaaccdcad8ec8eea2d7ba852574dbfc0b57956f5", "generation_seconds_batch": 18.12195374094881, "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": 6.4515180587768555, "genuineness": 4.595211505889893, "identity": 0.4505161941051483, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ada5638b6650fe5b4db6da63782ba0a415ca7f0da9400c41a04f5457c981d4ce.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "921f21f8dd704a3aeaf08499d0682dd9385d3667acdea8ed20098093814fba08", "generation_seconds_batch": 16.135575035004877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1029257774353027, "target_z": 1.7343750716161648, "ce": 6.054384231567383, "genuineness": 4.068024158477783, "identity": 0.6254193186759949, "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_high/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.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": "ade67da6a972b4d432baf7eb0e1e01c1c13d75972bb2c3174d7bc7bca5d5c612", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b0dd7bd5ffd4c0601b2ad83f615c3789c4dbaa8578aae429446d67aa1284776", "generation_seconds_batch": 21.495543014025316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.571082830429077, "target_z": 2.296875071616165, "ce": 5.824047088623047, "genuineness": 4.017816066741943, "identity": 0.5442699193954468, "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_high/ade67da6a972b4d432baf7eb0e1e01c1c13d75972bb2c3174d7bc7bca5d5c612.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1a203817a493e66ed328a30cab63a6ccdc1b6fc79b377aac929f9c1d1ad51cce", "generation_seconds_batch": 18.394503772025928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537235975265503, "target_z": 1.0546875358080825, "ce": 6.284721851348877, "genuineness": 3.886019229888916, "identity": 0.5581200122833252, "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_high/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "ae49e4a9afe3043dd6ad72130e578bf35f951ed66c90ac3965fbf023add13d12", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 130, "seconds_generated": 10.4, "mp3_sha256": "02a4f501575ec4e0dac6e579cca0edd218546c651c33d3cad3e6843b7a23a8f7", "generation_seconds_batch": 16.874864493031055, "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": 5.526196479797363, "genuineness": 4.519899368286133, "identity": 0.45159628987312317, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble.", "wer": 0.15789473684210525, "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_high/ae49e4a9afe3043dd6ad72130e578bf35f951ed66c90ac3965fbf023add13d12.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": "aeb86faba67b6ccff03e8ad210e5dbb38f127f2bb57b16a2a8c637f3716c0e81", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b42bb9370544dac1f8895b1ea181d6c25c6224f68016b89df82755cbaba178a", "generation_seconds_batch": 24.333027897053398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.559414386749268, "target_z": 3.484375071616165, "ce": 5.808161735534668, "genuineness": 3.848362922668457, "identity": 0.4192574620246887, "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_high/aeb86faba67b6ccff03e8ad210e5dbb38f127f2bb57b16a2a8c637f3716c0e81.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "aefa4be9f7e45daf216275e30a23b4d4b1bf2a8a199e51bb88ee0c76728e4afb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0ff1f33b466af9d191990de70d2f71e667995c6424f45b1d808bd9f1bc137918", "generation_seconds_batch": 24.00642042804975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5647534132003784, "target_z": -0.1137695827241185, "ce": 6.348263263702393, "genuineness": 4.570107460021973, "identity": 0.5509796738624573, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/aefa4be9f7e45daf216275e30a23b4d4b1bf2a8a199e51bb88ee0c76728e4afb.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "af0951ad2fa19d1ab4e6ef16927e1fa749372409ff509628b4d65f45501828d7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a8b3a01ad357e1557e16272d183b4b3e56b0341826fa25768d8f82c1b08a33be", "generation_seconds_batch": 16.874864493031055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7772934436798096, "target_z": 0.14160148640782483, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.7231221199035645, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/af0951ad2fa19d1ab4e6ef16927e1fa749372409ff509628b4d65f45501828d7.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": "af4cf8892205100604f61fa461f1b734ebf54c8bbfe576b46b77ae30a96bb14a", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "752606219390c10dfe43719b8d946cb73d6718015c68a01391edd4a2a5f717b8", "generation_seconds_batch": 25.696245218045078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4020261764526367, "target_z": 2.0937501432323296, "ce": 5.692992687225342, "genuineness": 4.18099308013916, "identity": 0.7626513242721558, "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_high/af4cf8892205100604f61fa461f1b734ebf54c8bbfe576b46b77ae30a96bb14a.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "af83d83bddd0f61bda0346ddbc10dfa580eaf2a76671f43a6ccc9ea83b06ca99", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "67291cc1f92d3d2189c899075137b20c5f918ec3d0dd0ee1ecf89ecb0008c18e", "generation_seconds_batch": 22.838779081008397, "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": 6.260893821716309, "genuineness": 4.68307638168335, "identity": 0.7760533094406128, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/af83d83bddd0f61bda0346ddbc10dfa580eaf2a76671f43a6ccc9ea83b06ca99.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "af97c3b95ac2f7e075f75866da1179a19e430f028843f9c190a6f3b7462a5f94", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ef82d183db77b857887de0cd055acac7c767a86b92d233ba4166718f4485b199", "generation_seconds_batch": 24.822336797020398, "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": 6.070269584655762, "genuineness": 4.519899368286133, "identity": 0.5692698955535889, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/af97c3b95ac2f7e075f75866da1179a19e430f028843f9c190a6f3b7462a5f94.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "aff912de9ef14955feae75feba1bb9ca748b7ba9f0f44e3a92831ef348cf9179", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "27387e2ea216a20f37d9744a16ed5aa8b8a9b52f4af839ef83bc831079d205d7", "generation_seconds_batch": 17.166892051929608, "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": 6.046441555023193, "genuineness": 4.419482707977295, "identity": 0.589611291885376, "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_high/aff912de9ef14955feae75feba1bb9ca748b7ba9f0f44e3a92831ef348cf9179.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "affd93af0c809d3a4083ce53ad7e5b22eb538c74af3c8ba2e6d2a7fc04c27755", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "17c8f8a0da7a5a6be288d8c059c1a642652155667cc499a20269142bb63c201e", "generation_seconds_batch": 16.943611936061643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1094279289245605, "target_z": 1.7421875358080825, "ce": 6.38003396987915, "genuineness": 4.074300289154053, "identity": 0.4646545350551605, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/affd93af0c809d3a4083ce53ad7e5b22eb538c74af3c8ba2e6d2a7fc04c27755.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": "b04155eb46aa05ba2cc5d8e6d88382ccebf0c543c12c847d56da66728403cf9b", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "306865dac0dc25da62d46963fd902858b511a25d4a9415d008f1d7c61728c70a", "generation_seconds_batch": 17.971176479943097, "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": 6.109982967376709, "genuineness": 4.394378662109375, "identity": 0.4312089681625366, "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_high/b04155eb46aa05ba2cc5d8e6d88382ccebf0c543c12c847d56da66728403cf9b.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "b05b8cf15dcd9b3c7af93c9e4cb93a665fb04d4f3253a79e2ebb15546318e9f2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "579bcddcff069307ba3c8563f65f6dcd695a6ddca9ce24418d5c2a194c724945", "generation_seconds_batch": 23.249505473067984, "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": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.4973898231983185, "hypothesis": "I found a little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b05b8cf15dcd9b3c7af93c9e4cb93a665fb04d4f3253a79e2ebb15546318e9f2.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "b0ed317dd4b6439d15e7d3e32832be3f6114875690231ba6de61c55a389a6fbd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "08f47c36c4392319ec88fb560a3c5a4f2ebf1b5348337bddfc35b18eca8e7762", "generation_seconds_batch": 18.119154921965674, "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": 6.0146708488464355, "genuineness": 4.369274139404297, "identity": 0.6518220901489258, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b0ed317dd4b6439d15e7d3e32832be3f6114875690231ba6de61c55a389a6fbd.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "b119e8f233ce5ced8542466564a0bba98214bcab8763cb43e6207beb0a15899d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "14a3ef3d63d418c455a0756fd00f2bfe81d9d2caabde261cf5d185cebf24f57f", "generation_seconds_batch": 20.694722806103528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.624436378479004, "target_z": 3.5625002864646595, "ce": 6.0146708488464355, "genuineness": 3.7416701316833496, "identity": 0.38593897223472595, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast dat 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_high/b119e8f233ce5ced8542466564a0bba98214bcab8763cb43e6207beb0a15899d.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "b14e154b5698781ccc2c110dfff10d1a8f2f42f1cbee1c4506eb348a22c37ad1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "32307be210a468a05336ad5b33f5bdca28f28acad7b31097cf9399f59edbae05", "generation_seconds_batch": 24.132291703950614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9988908767700195, "target_z": 1.6093750716161648, "ce": 6.324435234069824, "genuineness": 4.394378662109375, "identity": 0.5507878661155701, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b14e154b5698781ccc2c110dfff10d1a8f2f42f1cbee1c4506eb348a22c37ad1.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "b174c5638d4e94075538364b1e07f57d9a4cdb5460679397e235991eabf4db01", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c7469e1a3f3d682163fbeb0d88e4c2a24ae46e3613b3b8c3d92fb3b06593ca64", "generation_seconds_batch": 18.739524059928954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.670413851737976, "target_z": 0.013183533323860902, "ce": 6.157639026641846, "genuineness": 4.632867813110352, "identity": 0.5495749711990356, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b174c5638d4e94075538364b1e07f57d9a4cdb5460679397e235991eabf4db01.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": "b18399c6dcd5ef4eb12e0c71e9521bcbedd9b41d5ae8502f9647baeab54646bf", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70e0f8267728dd3174c152800c36f25e9db82631fbe4e22e270c06625325b293", "generation_seconds_batch": 23.042630922980607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4020261764526367, "target_z": 2.0937501432323296, "ce": 6.0305562019348145, "genuineness": 4.068024158477783, "identity": 0.5514615178108215, "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_high/b18399c6dcd5ef4eb12e0c71e9521bcbedd9b41d5ae8502f9647baeab54646bf.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "b202ea5a5d7d15756269be8bc4f9465819970c0e11054d708c19da8f3856c138", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f847d3251aadf0a1a67acc9fb6317d68143e949756b7939fc6304376aca0384c", "generation_seconds_batch": 21.800078919972293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.610095977783203, "target_z": 2.3437501432323296, "ce": 5.796247482299805, "genuineness": 4.218649387359619, "identity": 0.09105467796325684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b202ea5a5d7d15756269be8bc4f9465819970c0e11054d708c19da8f3856c138.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "b259b84173699981451bca713b5da1662c653baaa5903b2783ac2ddd751e795d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1c32aeea13976377b90be0bc3e266f1a8a87e95a3f33b0dcdfac3c2e5f3048e4", "generation_seconds_batch": 17.07032868207898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8558428287506104, "target_z": 1.4375, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.5637024641036987, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b259b84173699981451bca713b5da1662c653baaa5903b2783ac2ddd751e795d.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "b2b6cd8128c5ba13a1653af140d2808f81951c1681ba11ee35bcbdfe7d47ef1d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b694c30079107c0b2b1b8351e661a552daad4b505d0d26fddbcf5c64698941e0", "generation_seconds_batch": 25.212228736956604, "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": 6.205295085906982, "genuineness": 4.425758361816406, "identity": 0.47528761625289917, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b2b6cd8128c5ba13a1653af140d2808f81951c1681ba11ee35bcbdfe7d47ef1d.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "b2e3e7d0153062f7be3299a0a46003515f1934ba117953daf10cd8faf93695af", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5a5328c4c3a2f3d18ecb18d73d3700cea5f3ecb2691af53f058dd68ef72b07d5", "generation_seconds_batch": 23.464857763028704, "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": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.8224996328353882, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b2e3e7d0153062f7be3299a0a46003515f1934ba117953daf10cd8faf93695af.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "b3ca4102e8c7f1a96273d88236e3aa097ead7ca4423fe97957944779cdaace0c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5e25be79588476be8815df4d064a2d80c7b6000e346fa1046ad4d923550c65a1", "generation_seconds_batch": 23.88750126992818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.491720676422119, "target_z": 1.0, "ce": 6.308549880981445, "genuineness": 4.293961524963379, "identity": 0.5170666575431824, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b3ca4102e8c7f1a96273d88236e3aa097ead7ca4423fe97957944779cdaace0c.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "b3f0fd9a3ec3d04999634b155c57eb1bf4a201a1c28fde4736221158bb129e4e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f4ec1ccba86d680784b9e136be6d3f59b61a46ab6eee19dcb1758edbc33e8983", "generation_seconds_batch": 23.596837450983003, "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": 4.833701133728027, "identity": 0.8049250841140747, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b3f0fd9a3ec3d04999634b155c57eb1bf4a201a1c28fde4736221158bb129e4e.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "b4010b1c970f493a6120b6d8cef60dc0cf35e10dbba96500824053c10e2ebfca", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e6c4f9bcc36f1133545a44260d8945f85391f1a2939c919422511e73960a0a0f", "generation_seconds_batch": 17.012744573061354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.666553258895874, "target_z": 0.0085449543260747, "ce": 6.2767791748046875, "genuineness": 4.607763767242432, "identity": 0.6888585686683655, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b4010b1c970f493a6120b6d8cef60dc0cf35e10dbba96500824053c10e2ebfca.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "3f4446e3a9fcc784269d77722dbe0b4e50d77be49026a5c1fe4d6caf18734d7b", "generation_seconds_batch": 15.729415972949937, "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": 6.419747352600098, "genuineness": 4.720732688903809, "identity": 0.5760281085968018, "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_high/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "b5afe2734553aba0e4e4ef17c04c0e2aeeac97fc9f23aa488266ec2000733484", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "559eeaa45ae7a779662edaa224965650985a0826c265e6aeebc542d74028af32", "generation_seconds_batch": 23.995354049955495, "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": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.6345810890197754, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b5afe2734553aba0e4e4ef17c04c0e2aeeac97fc9f23aa488266ec2000733484.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": "b6056fb18fa70cbc1eadb030cc333dbc69af150e1f1e232d3ad5962b72d52512", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ef3f929ecea1dd1893aef752e70d7a24d361d6b2d1311d87c3908db8596eb96", "generation_seconds_batch": 22.117929962929338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.413694381713867, "target_z": 0.9062498567676703, "ce": 6.125868320465088, "genuineness": 4.419482707977295, "identity": 0.5174113512039185, "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_high/b6056fb18fa70cbc1eadb030cc333dbc69af150e1f1e232d3ad5962b72d52512.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "b63a9648c49a90edcdd736d892b38f61282de41e53303777b6bdeb29761c0fc5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "597cd6a626b7d7b42eb46c848f84fe8635b1d7423f93d380a10fc6ccee693d09", "generation_seconds_batch": 16.37857794004958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4787163734436035, "target_z": 0.9843750716161649, "ce": 6.149696350097656, "genuineness": 4.381826400756836, "identity": 0.763837993144989, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b63a9648c49a90edcdd736d892b38f61282de41e53303777b6bdeb29761c0fc5.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "b6785ebcca0a604a1fe0304d21187cade7f956f64f0864f3edfe5c2fc15d0387", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1ee4b32b68308ad6b70c0040e6d177446dbb7172fad03a6626947ea90af55a2d", "generation_seconds_batch": 16.828247010009363, "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": 6.006728172302246, "genuineness": 4.645420074462891, "identity": 0.8226871490478516, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b6785ebcca0a604a1fe0304d21187cade7f956f64f0864f3edfe5c2fc15d0387.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "b679a3c91ba5d95313eb4f6fc04e81b5ed2235fde544f7364b1616aebb21401d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7f625b9805b5c67f62931e41260321148c9dbd38bb0bab875852ec1d255dc248", "generation_seconds_batch": 23.709912082063965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4819674491882324, "target_z": 0.9882813037121236, "ce": 6.205295085906982, "genuineness": 4.532451152801514, "identity": 0.7325607538223267, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b679a3c91ba5d95313eb4f6fc04e81b5ed2235fde544f7364b1616aebb21401d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "b75f83a8cb511d18749e204afe2d71108a62658184e07782514bac1e22bc7c7f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c5113de57185af250be7f5e547369028e96af40e4c0c750fa0f5949ca25160c2", "generation_seconds_batch": 23.709912082063965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940371036529541, "target_z": 1.539062320959588, "ce": 6.1894097328186035, "genuineness": 4.2374773025512695, "identity": 0.5856190919876099, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen het samen inkijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b75f83a8cb511d18749e204afe2d71108a62658184e07782514bac1e22bc7c7f.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "b7ae4dab9de6cf973b9dc5c43bc8299004a579b474d4e9425eb506e6904a01b7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bf2c4cc6a61f6d19e69378aef78eae5728d66ee6e27fbaa95af4df99afc338f6", "generation_seconds_batch": 22.331282382947393, "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": 5.1052350997924805, "genuineness": 4.13078498840332, "identity": 0.27282699942588806, "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_high/b7ae4dab9de6cf973b9dc5c43bc8299004a579b474d4e9425eb506e6904a01b7.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "b7bd6bd1eff2c44ee7ce74e29604634462f2e1b5930c9094f14ce8e62147fcdf", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "38186fc5f8a95b03634db4df15b5e3bbab6e5b5c0adeed1176a7bbdb67abc725", "generation_seconds_batch": 25.37856071803253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.610095977783203, "target_z": 2.3437501432323296, "ce": 5.003965854644775, "genuineness": 4.099404335021973, "identity": 0.6810761094093323, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b7bd6bd1eff2c44ee7ce74e29604634462f2e1b5930c9094f14ce8e62147fcdf.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "b7cfb3840ba683bbba0d9974fd621807be21e1b5c9722340ac44c4afba77c802", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0f16ea45513625507da693727457b9e0731e97bb9cdb488d391a6158c7748d7c", "generation_seconds_batch": 16.063230513012968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.831170082092285, "target_z": 2.609375071616165, "ce": 5.919358730316162, "genuineness": 4.024092197418213, "identity": 0.6216217875480652, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b7cfb3840ba683bbba0d9974fd621807be21e1b5c9722340ac44c4afba77c802.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "b7e579aac4db4b8ca79baf4af89b9a49bc20e83d67e89243972b2f357738d2ba", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "649df6e7dbc8f69ef675cf46e9ff834282c1028052500d75804ca92c94479862", "generation_seconds_batch": 15.453905324917287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524231433868408, "target_z": 1.039062320959588, "ce": 6.24500846862793, "genuineness": 4.406930446624756, "identity": 0.6279972791671753, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b7e579aac4db4b8ca79baf4af89b9a49bc20e83d67e89243972b2f357738d2ba.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": "b80e62d4f9d4a158386f931a99d7b34d29b4f3854e2e4eace5e6bb9f80bfed47", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1fe91d020c84ad65a2bfc74789fd9107bfb0986cd40e44e5a4c9bfe7f7cc5cee", "generation_seconds_batch": 17.2268352949759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1419386863708496, "target_z": 1.7812498567676702, "ce": 6.316492557525635, "genuineness": 4.250029563903809, "identity": 0.4299764633178711, "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_high/b80e62d4f9d4a158386f931a99d7b34d29b4f3854e2e4eace5e6bb9f80bfed47.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "b811ffdf385287596241861ac24ab9c3d9b73f600d4e9dd3d1408264a9ced2c9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4eb30b33872191092b1f6568984d9a19fc2efd45f17b0ed9970a0cd30abc062e", "generation_seconds_batch": 18.020601939992048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.719992995262146, "target_z": 0.07275385925189182, "ce": 6.395919322967529, "genuineness": 4.193545341491699, "identity": 0.5839382410049438, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b811ffdf385287596241861ac24ab9c3d9b73f600d4e9dd3d1408264a9ced2c9.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "b8ceebfdef01337d6805e4db45141d409e604c57dc8dcb4b6d0d23cc61304dda", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "48cb94753bf1d257b6e0d4c1c18ac72c791aa15dcd06a99ee4b23e10f73cc754", "generation_seconds_batch": 16.73251497000456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.390357494354248, "target_z": 3.2812498567676704, "ce": 5.289902210235596, "genuineness": 3.7165660858154297, "identity": 0.3691125512123108, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b8ceebfdef01337d6805e4db45141d409e604c57dc8dcb4b6d0d23cc61304dda.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "b8e5dae8fe60fdbb543d7b60ea6cb5c8c66c2e02f85a08b2791c5e4cedbd6de3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4e4e29699b2c8788e89c3dc42ee4ccf92ad106af603b7d43017bb2daec0814c3", "generation_seconds_batch": 17.9043094139779, "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": 6.2767791748046875, "genuineness": 4.2751336097717285, "identity": 0.5339792966842651, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b8e5dae8fe60fdbb543d7b60ea6cb5c8c66c2e02f85a08b2791c5e4cedbd6de3.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "b95fc6db0dfec6746876822af74c4bedef473e4d41e21893978556e415707d2a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "419ad399bf44e480cd660e54784327ea208bd29f0d3c9b938db04287021b8c52", "generation_seconds_batch": 18.020601939992048, "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": 6.2767791748046875, "genuineness": 4.507347106933594, "identity": 0.5102530717849731, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b95fc6db0dfec6746876822af74c4bedef473e4d41e21893978556e415707d2a.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "b9807070c83e655974094729725489f11c8afde7faae6e5b254925f612692ef8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "8bc7d7d830f34998b75ed654b263cf6e0503dc261980a9c4521b41fb8b572610", "generation_seconds_batch": 22.62674724601675, "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": 6.197352409362793, "genuineness": 4.6203155517578125, "identity": 0.6183439493179321, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b9807070c83e655974094729725489f11c8afde7faae6e5b254925f612692ef8.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "b9b7959fe76ee5d7189762c99670269644a77b70da26d124aa5d3ab3abbd7fbb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "da39a7d76c19d17b4f27b5f0d104bf3b97d8199d9501fddc7636040c3bec91eb", "generation_seconds_batch": 22.33074641006533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6679755449295044, "target_z": 0.010253859251891812, "ce": 6.419747352600098, "genuineness": 4.268857479095459, "identity": 0.5970401763916016, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b9b7959fe76ee5d7189762c99670269644a77b70da26d124aa5d3ab3abbd7fbb.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "b9c0f8aa5741a772fe1e2a8f644e57829c56f3b30e546d0b0f6c1d5364e0aea2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9800e5eebac5cb3879ae928493cfdf620f8fbebaeab8b2e3942a84b04ac35cec", "generation_seconds_batch": 22.353658317937516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.572418689727783, "target_z": 3.5, "ce": 5.982900142669678, "genuineness": 4.042920112609863, "identity": 0.3786892890930176, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b9c0f8aa5741a772fe1e2a8f644e57829c56f3b30e546d0b0f6c1d5364e0aea2.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": "b9dae35d30d8e895ff556012048bad51dcd1768891e68f479bd3b3e2660f33d1", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9f616e194124c2cd65f3cbd540bb22fcc0b88c49aaa8aacb9219962655354e22", "generation_seconds_batch": 22.828968478017487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.096423625946045, "target_z": 1.7265626074242473, "ce": 6.284721851348877, "genuineness": 4.432034492492676, "identity": 0.47893598675727844, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/b9dae35d30d8e895ff556012048bad51dcd1768891e68f479bd3b3e2660f33d1.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "ba2a9659e7c2568bc41ba54ce442620b066179c3ded4ffea14b23d26b7e1bd97", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d615b1b0fbffadd45330d9a15b546506dc0710cc7020db4cebcb31c5d2d2e758", "generation_seconds_batch": 16.973177323001437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524231433868408, "target_z": 1.039062320959588, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.4682912230491638, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ba2a9659e7c2568bc41ba54ce442620b066179c3ded4ffea14b23d26b7e1bd97.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "ba5eab089443a048f71145d8ce221a29406c882d7e60c619d08e08f9a0a42ba7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "45b3ec0c6ab5d40e13bcd24986116e746b68a28fa02cc63f1a422234d92747a0", "generation_seconds_batch": 22.16238859493751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.078252792358398, "target_z": 2.9062498567676704, "ce": 6.038498878479004, "genuineness": 3.986435890197754, "identity": 0.37340348958969116, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ba5eab089443a048f71145d8ce221a29406c882d7e60c619d08e08f9a0a42ba7.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "4d3dcb884cc7903a2cafb490aca936793b35c7d6a6976749a4c63299fc620917", "generation_seconds_batch": 17.888495799037628, "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": 4.545003414154053, "identity": 0.6496779322624207, "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_high/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "baf5d15d0d262bdac383bb25087b2b18affc8e1230b3b37944f099ad0d3b1c46", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d60a5cd58229a4ba4dcde78069ba13bae3c78dbb04aed4d8cf84ff802db28412", "generation_seconds_batch": 24.581176512059756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1874542236328125, "target_z": 1.835937679040412, "ce": 5.1350202560424805, "genuineness": 4.218649387359619, "identity": 0.38050955533981323, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/baf5d15d0d262bdac383bb25087b2b18affc8e1230b3b37944f099ad0d3b1c46.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "bb24444bdf9876381dcee8a9dc80c59a85558f9c002c2adde57515142b1acb40", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b943d0d89a9ce0a6ada97138d983b5a7e4a873b9233d8e1c13cc12573ea0830b", "generation_seconds_batch": 23.959679967956617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0219380855560303, "target_z": 0.4355468839520206, "ce": 5.8875885009765625, "genuineness": 4.607763767242432, "identity": 0.6723089218139648, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bb24444bdf9876381dcee8a9dc80c59a85558f9c002c2adde57515142b1acb40.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "bbb2293877a53021a2c7443ef68121e2310b60e849949caddb2a81afd3323a47", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e8b44ab469049ce08a47f94ab105a43a044e36faf3db27f40bcbca20ed7dc185", "generation_seconds_batch": 23.202552985982038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.088585376739502, "target_z": 0.5156249283838351, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.47793859243392944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bbb2293877a53021a2c7443ef68121e2310b60e849949caddb2a81afd3323a47.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "bbea832183185f1fa377cf79bdb39dd93c3e5f035849f27bf97b5a5183e64b14", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "76d3a6e1e76d821f1e8fb1b9d072e7bb357659d3aab826499c25a3a19c711308", "generation_seconds_batch": 17.48771045799367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.814681053161621, "target_z": 0.1865234419760103, "ce": 5.927301406860352, "genuineness": 4.256305694580078, "identity": 0.799476146697998, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bbea832183185f1fa377cf79bdb39dd93c3e5f035849f27bf97b5a5183e64b14.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "bc921e572b8e7b237f51d0721cba407ebafa13c889ba90968220efe4cc425c41", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "2a12e4abd837ce0ba7d52ad37d7690d8c1d1d7c595bfeff67417f681c80a1f7a", "generation_seconds_batch": 21.646045641973615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4410390853881836, "target_z": 2.140624928383835, "ce": 5.673136234283447, "genuineness": 4.137060642242432, "identity": 0.5210302472114563, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bc921e572b8e7b237f51d0721cba407ebafa13c889ba90968220efe4cc425c41.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": "bcee57c675a064073bb2134ef7e45ee987a681c5b2b99b6cb2a56f9ee4a1d095", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "183024ad78aa1512d93c026976d1d3bb4df4b86736ac6b1868b0561ca0502710", "generation_seconds_batch": 17.2268352949759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6997904777526855, "target_z": 1.25, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.24065202474594116, "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_high/bcee57c675a064073bb2134ef7e45ee987a681c5b2b99b6cb2a56f9ee4a1d095.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "bcf01e0ab09bf625d8a8486e61ea039ebd9b302f929087567933a7b94a4bc44e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ab59f6f7f1a33a6a702c5f7aff9273c4fccd92eef2ac0d4e4833b5701e88dee3", "generation_seconds_batch": 23.825047438964248, "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": 6.340320587158203, "genuineness": 4.582659721374512, "identity": 0.6102518439292908, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bcf01e0ab09bf625d8a8486e61ea039ebd9b302f929087567933a7b94a4bc44e.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "bd764387d902cb3d6f404eb585b7d9851da66064d895ca1859f95f25bd9688d2", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ea6e50b669cd45e2eee79d998420e6bb36111389f1b3ad90d54bc15bd882759f", "generation_seconds_batch": 16.845726281986572, "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": 5.796247482299805, "genuineness": 4.218649387359619, "identity": 0.5158259868621826, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bd764387d902cb3d6f404eb585b7d9851da66064d895ca1859f95f25bd9688d2.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "bda799706914cdbda2b877d442cb23b33340bddf29a186de542a7cd54ce24aad", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "10c068f493e90de8be66bbb2288878b90e375816b9591277d0d180ec6de9c07a", "generation_seconds_batch": 16.861858085962012, "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": 6.09409761428833, "genuineness": 4.532451152801514, "identity": 0.7437183260917664, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bda799706914cdbda2b877d442cb23b33340bddf29a186de542a7cd54ce24aad.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "be4df5b1d80ac04a88844369fe16ab6d022da9e6a317ed1e214557d767a29ae2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b0101d82e16ea77a18f162fccc1a5aeb1bb421e31e12945adc8b3e886cb4ce46", "generation_seconds_batch": 22.28453195595648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.170619487762451, "target_z": 4.21875014323233, "ce": 5.073464393615723, "genuineness": 3.4247303009033203, "identity": 0.5015018582344055, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/be4df5b1d80ac04a88844369fe16ab6d022da9e6a317ed1e214557d767a29ae2.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "be59d2482ebb0fe3a1bae4179246c91f96f5bf3108adb52a4bae25b9fa53f186", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5d7a00f7eb14a7676148fc5f49b578429ed22c77e1924902c8a8afaf4dfe5c03", "generation_seconds_batch": 16.987020627013408, "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": 6.205295085906982, "genuineness": 4.821148872375488, "identity": 0.5318865776062012, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/be59d2482ebb0fe3a1bae4179246c91f96f5bf3108adb52a4bae25b9fa53f186.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "be82371db8a019574104f17075010bd1d3e96b69ed009faebd7b60cc2511edc8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3d71a945e638d212965511840c702dc46139921d61ad3758f14b83c97c8519bf", "generation_seconds_batch": 22.42010622296948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.702462196350098, "target_z": 3.6562498567676704, "ce": 5.174733638763428, "genuineness": 3.280381202697754, "identity": 0.31750908493995667, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/be82371db8a019574104f17075010bd1d3e96b69ed009faebd7b60cc2511edc8.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "beb832307f34336a42b27c1dbebfb05c08f81d006701f58c5f4d3b38dd5b17f5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "af89a00093d09911a68e23f76ffbea1efe2e69ac8d768b00fa0087a95043c669", "generation_seconds_batch": 22.969574319082312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7648122310638428, "target_z": 1.3281249283838352, "ce": 5.998785495758057, "genuineness": 4.325342178344727, "identity": 0.480010449886322, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/beb832307f34336a42b27c1dbebfb05c08f81d006701f58c5f4d3b38dd5b17f5.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "bedf7cef4630db0de176d4bfda35d30c2a7174f19dff1b65ec15542bdafa6dc9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 129, "seconds_generated": 10.32, "mp3_sha256": "7dbdce939771fe56fab2ee1bece7bc3f9cdda495e9ddfa8f30fae8d7621efb7a", "generation_seconds_batch": 23.123347540968098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467047929763794, "target_z": 2.171875071616165, "ce": 5.283945083618164, "genuineness": 4.168440818786621, "identity": 0.3583011031150818, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bedf7cef4630db0de176d4bfda35d30c2a7174f19dff1b65ec15542bdafa6dc9.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "bf70cc3dd5a996f551193bbcc576bde67c1b068a8e0483147ab7bd4ac7c08c11", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "61d59fe3140132dfad3cb24c26e8c114730203398e42df657fc933212d6557a3", "generation_seconds_batch": 22.909973556990735, "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": 6.395919322967529, "genuineness": 4.557555198669434, "identity": 0.7745492458343506, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bf70cc3dd5a996f551193bbcc576bde67c1b068a8e0483147ab7bd4ac7c08c11.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "bff94976b359ef1f8dca364f86e310b910433848ac0329110a76585f2fa1c505", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "249f8c1e4751284c304375551e06c3228f848655a6843522eb96b1f7b94ae7c4", "generation_seconds_batch": 18.30825455102604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5001380443572998, "target_z": -0.19140623209595878, "ce": 6.157639026641846, "genuineness": 4.7834930419921875, "identity": 0.6830222010612488, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/bff94976b359ef1f8dca364f86e310b910433848ac0329110a76585f2fa1c505.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "c062b536ccb904529af77fa48c8eab64f5c3345ee5210e88ec4d10fdef175b1d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "13373a10c3803c78c328c60d36fa04762a0731dfe02eef2a34e76e2b6e742e43", "generation_seconds_batch": 15.745475627947599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3291661739349365, "target_z": 0.8046875358080824, "ce": 6.2767791748046875, "genuineness": 4.068024158477783, "identity": 0.30069833993911743, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c062b536ccb904529af77fa48c8eab64f5c3345ee5210e88ec4d10fdef175b1d.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "c1075d986f6efa4b3e11d454bbaa01f70080198377f9236ae31d359f3f2546ff", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f4e9b1e5b0da95e8f70d9537d72fd72b0a4a4bee5d6aa2061f5eeb818b1718da", "generation_seconds_batch": 16.91786885797046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.452707529067993, "target_z": 0.9531249283838351, "ce": 6.157639026641846, "genuineness": 3.8797430992126465, "identity": 0.33502352237701416, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c1075d986f6efa4b3e11d454bbaa01f70080198377f9236ae31d359f3f2546ff.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c16067442a33f9dda0cb413110cd694d430574198137047a01d7eb23ba72a033", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2676da2dfa5a53a64d923d6feb237c91425f0c841300fff8fe34f1e69994fd8d", "generation_seconds_batch": 16.627182731986977, "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": 5.847874641418457, "genuineness": 4.256305694580078, "identity": 0.5486268401145935, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c16067442a33f9dda0cb413110cd694d430574198137047a01d7eb23ba72a033.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "c170ebee92609ca8f50e09e9c3e7cf465ea08323ff6fb682739c858ed12b1bee", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5833b948a0f7c8a8c6cdf30149a71d90620827b483c47dd4e55662622fe03ba2", "generation_seconds_batch": 15.514108174014837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6240859031677246, "target_z": -0.04248041727588151, "ce": 6.165581703186035, "genuineness": 4.7081804275512695, "identity": 0.6253363490104675, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c170ebee92609ca8f50e09e9c3e7cf465ea08323ff6fb682739c858ed12b1bee.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c20db8b6b5de26027887c68f3ae6a82dfea23f91bbffffac23c79830f42c4256", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f70a0a94f79af84d7097e7f537eee0df0423ad52c1de20dbb252ea0010d5317", "generation_seconds_batch": 24.214766227989458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9988908767700195, "target_z": 1.6093750716161648, "ce": 5.597680568695068, "genuineness": 4.13078498840332, "identity": 0.40094655752182007, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c20db8b6b5de26027887c68f3ae6a82dfea23f91bbffffac23c79830f42c4256.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "c22e1f068c1179e29bb6ffe0284af476c76bad34d61ceb8959e5a8c277e9e5c6", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fc0fda22287aeea68106459b87671ac0285b82726a5be5985a426dc5cb889df8", "generation_seconds_batch": 16.22692346200347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5177292823791504, "target_z": 1.0312498567676702, "ce": 6.054384231567383, "genuineness": 4.287685871124268, "identity": 0.6620141267776489, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c22e1f068c1179e29bb6ffe0284af476c76bad34d61ceb8959e5a8c277e9e5c6.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": "c27b0ad55b7e724dba31de77c9854dc70d1fbcd95e1a701b2a0104db9f827dc7", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "97c770ee03392fe80005c6c6c56544c70943f8a513fd321f64c42fcf74d4841e", "generation_seconds_batch": 16.434750373940915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0639126300811768, "target_z": 1.6875, "ce": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.5880308151245117, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería. a la estantería alta. La colocaremos después de comer.", "wer": 0.15789473684210525, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c27b0ad55b7e724dba31de77c9854dc70d1fbcd95e1a701b2a0104db9f827dc7.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": "c28113f40250eb64550435234de62c5e00a9c164257c0ff64f3fa4977bcb1318", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ad0ff09c7d2474e1c33d54b32994ad46f8cbd97342bae2372bd3c11156451bc4", "generation_seconds_batch": 17.969894291018136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1809518337249756, "target_z": 1.8281249283838352, "ce": 6.348263263702393, "genuineness": 4.243753433227539, "identity": 0.4785863161087036, "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": 0.120000000000001, "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_high/c28113f40250eb64550435234de62c5e00a9c164257c0ff64f3fa4977bcb1318.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": "c2ccd9d3bd47a943af8783a4c405ef598ecfc8ff92f2e773b0d1cd3108631bc8", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76a7df3d6dee687d504f587875acbe2d98edc0d6fdfa6dadc24e04867c19be67", "generation_seconds_batch": 23.135435128002428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.452707529067993, "target_z": 0.9531249283838351, "ce": 5.736677646636963, "genuineness": 4.457139015197754, "identity": 0.7642549276351929, "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_high/c2ccd9d3bd47a943af8783a4c405ef598ecfc8ff92f2e773b0d1cd3108631bc8.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "c379c81cf8aad99dc1218c0d2bc51200bd69d545756ca26945cbb0a28342baaa", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a5c2fd7ca81d8be54f8a7948ff7d48fcca7c6d967f47119cd9da4fe0246c5fd2", "generation_seconds_batch": 17.57903288397938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5464661121368408, "target_z": -0.13574213826388667, "ce": 6.09409761428833, "genuineness": 4.432034492492676, "identity": 0.7392625212669373, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c379c81cf8aad99dc1218c0d2bc51200bd69d545756ca26945cbb0a28342baaa.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d106146aca665bbbe6e3f85f737eb1375b96201234907e06be719253fc93b199", "generation_seconds_batch": 16.857097771018744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.465712070465088, "target_z": 0.9687501432323297, "ce": 6.054384231567383, "genuineness": 4.262581825256348, "identity": 0.6956627368927002, "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_high/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "c3ec447c496fbcc53abbc2043342048af3d2b1025f1df7d6ae2898bd14338090", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "70ce14b64e2efae7f15aba29c1a330c097972273fdd89dc78a3aa01b68d1f2e4", "generation_seconds_batch": 24.068337997887284, "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": 6.054384231567383, "genuineness": 4.796044826507568, "identity": 0.5629183053970337, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c3ec447c496fbcc53abbc2043342048af3d2b1025f1df7d6ae2898bd14338090.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "c442ee30996d0293cfad8d6b2567199caf5b7f3b409a6e656ef6408afd62149c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "0af07b8ced2221491f1377595ae0a73e6bc9b66c6102b2a886ac30ea504da65e", "generation_seconds_batch": 18.11708835104946, "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": 6.324435234069824, "genuineness": 4.833701133728027, "identity": 0.6219555139541626, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, eh. 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_high/c442ee30996d0293cfad8d6b2567199caf5b7f3b409a6e656ef6408afd62149c.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "c465a71fbdb0870a5ed37fd59649a0019c325aac96cc6fe7b2b52491e99cd158", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "58b006e0234c1f811eca01ab919d95c739996e2285c127f2c4a69747fdbc7fda", "generation_seconds_batch": 22.945971267065033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9585418701171875, "target_z": 0.35937507161616483, "ce": 6.165581703186035, "genuineness": 4.68307638168335, "identity": 0.7029560804367065, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c465a71fbdb0870a5ed37fd59649a0019c325aac96cc6fe7b2b52491e99cd158.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff42a69c4cda4a34be6250a96d836c0f68c5e526fa767906aa4910c6c494d283", "generation_seconds_batch": 17.855580860050395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892534255981445, "target_z": 1.1171875358080825, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.5982633829116821, "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_high/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "c4d667d31d7f3a1c9780219e21fe97c746c099a352a5551757541de3adcf8927", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b5999475a73310d2e5b19540f6545e8f709b5f59201aadaa8924c124aa31b32c", "generation_seconds_batch": 22.89164546399843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7423442602157593, "target_z": 0.09960934814393818, "ce": 6.23706579208374, "genuineness": 3.8922953605651855, "identity": 0.22357846796512604, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c4d667d31d7f3a1c9780219e21fe97c746c099a352a5551757541de3adcf8927.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "c4deb30c44a548c2a7988c3420ec97c70b4c8581ae2db6d600c1217b28e14e6c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4974dc6d038680cdc66bd61af1dd98d9798263dbe2e0ff4a920e674947404909", "generation_seconds_batch": 18.26535903196782, "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": 6.316492557525635, "genuineness": 4.645420074462891, "identity": 0.4861302375793457, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c4deb30c44a548c2a7988c3420ec97c70b4c8581ae2db6d600c1217b28e14e6c.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": "c513742492aae88aafd12e6a3421a2a174d1d5034eef156e359361ae16cd8a82", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ff848d96e935daa7b8bf2ac7a2c913da429c092853fd3b06c04a56d344c9984", "generation_seconds_batch": 18.423899870947935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.576249122619629, "target_z": 1.1015626074242473, "ce": 6.09409761428833, "genuineness": 4.400654315948486, "identity": 0.6695382595062256, "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_high/c513742492aae88aafd12e6a3421a2a174d1d5034eef156e359361ae16cd8a82.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": "c52e34bc7d82e127a7c787d979ffd2fad876dfdb3da1493ffddd84dd5d904c6e", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5e16f018cb873e051c5b3f7e10c63dbc1e8de8ab736ac7600cc9bdc1efaca934", "generation_seconds_batch": 22.21040278195869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8831875324249268, "target_z": 2.671875071616165, "ce": 5.911416053771973, "genuineness": 3.9927117824554443, "identity": 0.7473593950271606, "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_high/c52e34bc7d82e127a7c787d979ffd2fad876dfdb3da1493ffddd84dd5d904c6e.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "c5492cc51c8eee5a123e8f2d71b3900f3777bf6bcac197df8212b9a7f87bcef8", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ecf054b880d1741381af6665ef734a1cf77d6807c0ef680a4f062fa4ab9858e3", "generation_seconds_batch": 17.258045624941587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4071922302246094, "target_z": 0.8984373925757527, "ce": 6.229123115539551, "genuineness": 3.9676074981689453, "identity": 0.4725034236907959, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c5492cc51c8eee5a123e8f2d71b3900f3777bf6bcac197df8212b9a7f87bcef8.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": "c57c8326b1da14386431bdcf3095878e55231b252ff35ac1c0d18a5d572406b9", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9eae5fd52cf74f4a8170b70e31f55769d9dbf542f605b134555244193f9999f", "generation_seconds_batch": 23.555441918899305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0704147815704346, "target_z": 1.6953124641919175, "ce": 6.340320587158203, "genuineness": 3.9299514293670654, "identity": 0.3889949917793274, "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_high/c57c8326b1da14386431bdcf3095878e55231b252ff35ac1c0d18a5d572406b9.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "c59bda433062faa7c51c44c6d915ae0fb369a80410d314aed5c75badc8719064", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "61e892683aa65c3be2f019ad7f5a545e6f57b6c0d56bd343b572d8d69d5cf92d", "generation_seconds_batch": 17.42719495401252, "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": 6.443574905395508, "genuineness": 4.607763767242432, "identity": 0.49178406596183777, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c59bda433062faa7c51c44c6d915ae0fb369a80410d314aed5c75badc8719064.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "c5c8c1d90cc99d7353cac81dcfff52912c77b0e6267cd3c68e7454f04cb0568f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5563255a46c66ca725f871edc2a5072d226690e054c172dd98bbc2bd0df7f0e8", "generation_seconds_batch": 23.166388290934265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.390936851501465, "target_z": 0.8789062320959588, "ce": 6.0146708488464355, "genuineness": 4.607763767242432, "identity": 0.5792780518531799, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c5c8c1d90cc99d7353cac81dcfff52912c77b0e6267cd3c68e7454f04cb0568f.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c5e17b88b3179ebf8b913321f6fb303ff11741bd015050cdcf752fd19597abb9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8ec51e6af03dc6b94b8e8f431cc4942b9252c1af63adae05ca9f137244e4fed3", "generation_seconds_batch": 22.97839677601587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3383402824401855, "target_z": 3.2187501432323296, "ce": 5.462655067443848, "genuineness": 3.9048471450805664, "identity": 0.6033141016960144, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c5e17b88b3179ebf8b913321f6fb303ff11741bd015050cdcf752fd19597abb9.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "c5fe33093af320be9e8b3c38e63a9d4f59c326961e150f9b579b068ffdd82e8d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bc08a3510368288fe840b2a42b5c7a68d1cce4ab8c5e43680610fd41043fcef6", "generation_seconds_batch": 22.003199038095772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5871046781539917, "target_z": -0.08691409383207213, "ce": 6.133810997009277, "genuineness": 4.645420074462891, "identity": 0.7078368067741394, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c5fe33093af320be9e8b3c38e63a9d4f59c326961e150f9b579b068ffdd82e8d.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "c6639355f32d9bd3366c43799f6d05f82cc2cf3171d60ca005b90e2eb5229c9e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "65b8a5c574293d094ec07f2c5f89708e65e726de62a1e233e4c1e0dafe52c473", "generation_seconds_batch": 16.614855902036652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.65029776096344, "target_z": -0.010986349386048942, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.7090800404548645, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c6639355f32d9bd3366c43799f6d05f82cc2cf3171d60ca005b90e2eb5229c9e.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c66d2e7bf76e5110c746dfb16fbf235255d6871e74693f96e3fb5852067c1c34", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0f8d315f1b2ad1b90834b750e7107146b3d697747115f1065e2804f761412f97", "generation_seconds_batch": 22.035793457063846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.454043388366699, "target_z": 2.1562498567676704, "ce": 6.0146708488464355, "genuineness": 4.149612903594971, "identity": 0.7057678699493408, "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_high/c66d2e7bf76e5110c746dfb16fbf235255d6871e74693f96e3fb5852067c1c34.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "c697346cdacb377b85ad58a78821c4bba419d5834dd161d236842ec04cf293ae", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "22f37ecaf702cba7d9841613a31d3d411392b7992f9bfc23d8927a6b345055ea", "generation_seconds_batch": 15.622364434995688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.310995578765869, "target_z": 1.9843750716161648, "ce": 6.252951145172119, "genuineness": 4.168440818786621, "identity": 0.5255351066589355, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c697346cdacb377b85ad58a78821c4bba419d5834dd161d236842ec04cf293ae.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "c7017784b1d22870098735a45e50e407a1aa819fb4936b6f6ea8ad012293b395", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 129, "seconds_generated": 10.32, "mp3_sha256": "933398569549af54384403135e41053bfe083d7a9a50eb89e608ac729a0d0f3a", "generation_seconds_batch": 21.599320476991124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.797323226928711, "target_z": 1.3671875358080825, "ce": 5.4825119972229, "genuineness": 4.425758361816406, "identity": 0.5915896892547607, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "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_high/c7017784b1d22870098735a45e50e407a1aa819fb4936b6f6ea8ad012293b395.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "c7555cdbadb4096b7ee776f5175c39705ad8b325f87e7dd17fe95f7899c9c6dd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6dd9051091df55468220e839bdec085c0b1e8fb15ea2586a44e35ceec1f4b691", "generation_seconds_batch": 23.995354049955495, "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": 6.459460258483887, "genuineness": 4.6203155517578125, "identity": 0.5380659103393555, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c7555cdbadb4096b7ee776f5175c39705ad8b325f87e7dd17fe95f7899c9c6dd.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "c770a5d3c736af90608aa9d4e17ec45f84a75d37b70c15e70316d25682012f8d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1d2b9089c8673a3609d48d28ca6c9414b1d668ada4c5906cfa77c6c08822a20b", "generation_seconds_batch": 22.683026877930388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3630130290985107, "target_z": 2.046875071616165, "ce": 5.879645347595215, "genuineness": 4.362998008728027, "identity": 0.6967428922653198, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c770a5d3c736af90608aa9d4e17ec45f84a75d37b70c15e70316d25682012f8d.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c7bec8fa30496171f48671b0d7340932bd60fd1d14100b61f3c826347ba1ccda", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e1082697257b1f2fca0288440597cdcfb4a7f390e878b406047d03e2870a11df", "generation_seconds_batch": 22.331282382947393, "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": 5.855817794799805, "genuineness": 4.13078498840332, "identity": 0.4878762662410736, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c7bec8fa30496171f48671b0d7340932bd60fd1d14100b61f3c826347ba1ccda.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "2b416b0a7029ae2558c50b2ee5221674ecfeed4902cfc8943f88ae22c0f7d60d", "generation_seconds_batch": 16.135575035004877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6542751789093018, "target_z": 1.1953124641919175, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.6904774308204651, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las de 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_high/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "c867a9beb0068f354ebfc5d8fd9ef03d869fb3a77093f11b1e9e615427862241", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d6632520e7572842dcba1f8e76752948a65f8ec39ff1ce09fa762ab2bed39c0e", "generation_seconds_batch": 21.88348264200613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.546410083770752, "target_z": 3.4687501432323296, "ce": 6.1020402908325195, "genuineness": 3.860914945602417, "identity": 0.4472739100456238, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c867a9beb0068f354ebfc5d8fd9ef03d869fb3a77093f11b1e9e615427862241.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "c870ad3928e97549f61592278433e8130d8b0a94bbecc23103bb8da88c8a5554", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "20596dfe742d5a9afe47496f2d9f83e6449d734fca9e8085ab3d8f6af6df188a", "generation_seconds_batch": 23.123347540968098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0314016342163086, "target_z": 1.6484373925757527, "ce": 5.442798614501953, "genuineness": 4.356721878051758, "identity": 0.4432064890861511, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c870ad3928e97549f61592278433e8130d8b0a94bbecc23103bb8da88c8a5554.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "154487dade3885e73b999c2870e49c1dc1ec804b79ead1ecfdeed8c811167426", "generation_seconds_batch": 23.290208989055827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1145942211151123, "target_z": 0.5468750716161649, "ce": 6.4515180587768555, "genuineness": 4.243753433227539, "identity": 0.4045836627483368, "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_high/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "c8fed67ae1981a6605960b1796d5ef565b4ec1982ed1d9c968ae3066ae76fa0e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b7226ac3a88ff14a9004ea2dd31c6ed2b7db863552c704b06a88e98fc5a74d2d", "generation_seconds_batch": 24.17070268804673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2914891242980957, "target_z": 1.960937679040412, "ce": 6.0305562019348145, "genuineness": 4.168440818786621, "identity": 0.43046125769615173, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c8fed67ae1981a6605960b1796d5ef565b4ec1982ed1d9c968ae3066ae76fa0e.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "c914f243fb8683be102263f304c282b4783da09b43658a50bc9a7f1a617a8ed5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "35b9baeac8bc26c19582242d632597c140d7bffd56b8720c815813de245b4f3c", "generation_seconds_batch": 18.12195374094881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8138682842254639, "target_z": 0.1855468839520206, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.5258679389953613, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c914f243fb8683be102263f304c282b4783da09b43658a50bc9a7f1a617a8ed5.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "c97278efbfb08298bffedad2c2701d8fa12553ba70db42e8dc7fe63a8eb57ba8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "274e79b65ae353680f262124023b74cbc39c859461ef31c24ec07ecd6d520dd8", "generation_seconds_batch": 22.6588852860732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.870182991027832, "target_z": 2.6562498567676704, "ce": 5.454712867736816, "genuineness": 3.9048471450805664, "identity": 0.6239871978759766, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c97278efbfb08298bffedad2c2701d8fa12553ba70db42e8dc7fe63a8eb57ba8.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "c9b02246c44e72bd5d248aad1eff9a45b76e77c869101db793ca2f6837634d01", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b8f8e220a561fddc71d4c7ed16fa0b6bdf7142b04cd28d6f40d369d6867372a3", "generation_seconds_batch": 25.59293839603197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.389021635055542, "target_z": 2.078124928383835, "ce": 5.903473377227783, "genuineness": 4.306513786315918, "identity": 0.6134657859802246, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c9b02246c44e72bd5d248aad1eff9a45b76e77c869101db793ca2f6837634d01.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "c9c7159b284b40b5a114ae893c53a57adf2bf81b405b94561de0b5d65b36c824", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f6836bce016a13f8fa77717504456664824081f09e63598dfac0f9e538801067", "generation_seconds_batch": 22.909973556990735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1438539028167725, "target_z": 0.582031160479794, "ce": 6.260893821716309, "genuineness": 4.469690799713135, "identity": 0.5148986577987671, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie standen neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/c9c7159b284b40b5a114ae893c53a57adf2bf81b405b94561de0b5d65b36c824.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "ca0effc9f127f0742d95157facd3ff9300caaf3192ebe2f7997517810de78765", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5e20a45681f9b3a4f99cdb7391128ba86a97350dad66646f52a10846979db022", "generation_seconds_batch": 16.671934792073444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892534255981445, "target_z": 1.1171875358080825, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.7454169988632202, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ca0effc9f127f0742d95157facd3ff9300caaf3192ebe2f7997517810de78765.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "ca5e5e29c176c79554c0636508f66225937aad5d3758c7587c1943f0a6fdad1d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "23cc9852a170757730dda41b8b7f273ebe9238026370106d3dd747787a901a19", "generation_seconds_batch": 24.126250863075256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6431859731674194, "target_z": -0.01953130371212364, "ce": 5.685050010681152, "genuineness": 4.432034492492676, "identity": 0.7396239042282104, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ca5e5e29c176c79554c0636508f66225937aad5d3758c7587c1943f0a6fdad1d.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": "cab0b87f73f6a6878a8d387ec4da7f5f90582c94257118eb83e05b914ae6b520", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d204be49277f179f8ba2a0252ce4d500023c691c695576324ee13a3484c630b", "generation_seconds_batch": 23.030612318078056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.946873426437378, "target_z": 1.5468750716161648, "ce": 5.847874641418457, "genuineness": 4.545003414154053, "identity": 0.6103211641311646, "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_high/cab0b87f73f6a6878a8d387ec4da7f5f90582c94257118eb83e05b914ae6b520.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "cada9faf5a47f55e042f1b0136bbc911c3c646aedbaabe51d2639de09d929a30", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "76c1196d8fb1ea27107341c201a1893391dbfd28c51040335afd4536726f76c8", "generation_seconds_batch": 23.025358312996104, "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": 6.157639026641846, "genuineness": 4.3127899169921875, "identity": 0.6254235506057739, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cada9faf5a47f55e042f1b0136bbc911c3c646aedbaabe51d2639de09d929a30.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "caf154d4832304ca24ba7450eaae820e8990c4cabd89dd41cf7eba62be9aa8a0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "56aa27fd7fbb059dbc7f04d067ef490c42c13506c94f67e08bc08d9077830ba4", "generation_seconds_batch": 22.35609524208121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.454043388366699, "target_z": 2.1562498567676704, "ce": 6.268836498260498, "genuineness": 3.9801597595214844, "identity": 0.48540523648262024, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/caf154d4832304ca24ba7450eaae820e8990c4cabd89dd41cf7eba62be9aa8a0.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "cb1e573a6726c3fd9436ec0e248d766aaf583a0f46c4cf7f793699d33b99db4f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "08c939b15adce42ce5f6542add86180ab1c7d74388e3c22813cae15da70ff3fb", "generation_seconds_batch": 22.85636467509903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.096423625946045, "target_z": 1.7265626074242473, "ce": 5.450741291046143, "genuineness": 4.105680465698242, "identity": 0.4513906240463257, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cb1e573a6726c3fd9436ec0e248d766aaf583a0f46c4cf7f793699d33b99db4f.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "cb4562479a33c5e6119018b69960c7ded626fd720ae096d59d23ac676ea236b5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "19e580a9f102a0e83cb0e669660ff8954ba9670e7ba12c9c9f3c8e873acdf6b3", "generation_seconds_batch": 25.05968925799243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6830118894577026, "target_z": 0.02832032592803091, "ce": 6.213237762451172, "genuineness": 4.570107460021973, "identity": 0.7126888036727905, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cb4562479a33c5e6119018b69960c7ded626fd720ae096d59d23ac676ea236b5.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "cb8ac05a45b48a5f0b4ed855e04607be958cfa8cadf3e5cc8b6ef3c956aae7c9", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "816d9fa9932ea50209406b8e6b5368d629c3b64d949b77e0d5d6341616f6efa0", "generation_seconds_batch": 18.25526084995363, "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": 6.062326908111572, "genuineness": 4.13078498840332, "identity": 0.44225990772247314, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cb8ac05a45b48a5f0b4ed855e04607be958cfa8cadf3e5cc8b6ef3c956aae7c9.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "cb8f02ac399eb7eee2b8a0075e5600c8d2931ccf5f8694a7e5bd90065133fa42", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "01f004d71205a2ae2c042657f9aac29380efccaf5ff95e0da67bec0c19a6c87e", "generation_seconds_batch": 25.02847490296699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 5.7406487464904785, "genuineness": 4.507347106933594, "identity": 0.5410512089729309, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cb8f02ac399eb7eee2b8a0075e5600c8d2931ccf5f8694a7e5bd90065133fa42.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "cc3a1d8afd93fccd05d1fd0a956d31bd600565bc430889e761b5772f167640d5", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d8eac4936ce51f763fecef75daf8970e4d67b67bdd171f21200b5173f3cdec3b", "generation_seconds_batch": 17.303496812004596, "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": 5.903473377227783, "genuineness": 4.331617832183838, "identity": 0.6661540269851685, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cc3a1d8afd93fccd05d1fd0a956d31bd600565bc430889e761b5772f167640d5.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": "cc6c1ea5417e5282b1af521b7c0619d1f2078222aca8548d8cf63b1e5e829b51", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ff7bfca605cf0b15d0f21d01bea69d63e8bad27fa1c1a1f5047edbb468d9ebe3", "generation_seconds_batch": 23.385942174936645, "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": 6.197352409362793, "genuineness": 3.8044304847717285, "identity": 0.426596462726593, "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": 0.120000000000001, "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_high/cc6c1ea5417e5282b1af521b7c0619d1f2078222aca8548d8cf63b1e5e829b51.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "cd285824b37269ecb205ca294ea61b5c1afe39cbde0cd32ee44b7a2ba700abfe", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "959ec6223e90fa7b007a117502ecfdca5236aeeaf6c19169b43232577b8310cb", "generation_seconds_batch": 24.000451322994195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.024899482727051, "target_z": 1.6406249283838352, "ce": 5.125091552734375, "genuineness": 3.9173994064331055, "identity": 0.47910022735595703, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cd285824b37269ecb205ca294ea61b5c1afe39cbde0cd32ee44b7a2ba700abfe.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": "cdb3a21c211d5d379a10d279bac52ab665e6d6aef6123895bbec360380493a9d", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2131834038ff9c13b5572fdc711994fd0ced9d2412cb4a13a91d35bb99ea545c", "generation_seconds_batch": 22.89692713401746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0053930282592773, "target_z": 1.6171875358080825, "ce": 5.982900142669678, "genuineness": 4.256305694580078, "identity": 0.7997187376022339, "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_high/cdb3a21c211d5d379a10d279bac52ab665e6d6aef6123895bbec360380493a9d.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "ce0b7399c096f8c3ac50d6f442e4c1bfec18ac129d750296a866c2b33488c0ec", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "32a8af79d53ed3b0df455fb34d71093b1bd47216f7e84d48fc7be1dc6e0abd5d", "generation_seconds_batch": 25.53965748997871, "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": 6.078212261199951, "genuineness": 4.068024158477783, "identity": 0.589880108833313, "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_high/ce0b7399c096f8c3ac50d6f442e4c1bfec18ac129d750296a866c2b33488c0ec.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "ce98d2465e6216359b7a6863b976b2aaf313f6559b78b99cb4cbac89fdfea70d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1bd10b4ce42f97048b707eed1a6959a3d3f78ca47ebf25e08a22f8d30ec254da", "generation_seconds_batch": 16.489943793974817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6217641830444336, "target_z": 1.1562498567676702, "ce": 6.205295085906982, "genuineness": 4.030367851257324, "identity": 0.3641415536403656, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ce98d2465e6216359b7a6863b976b2aaf313f6559b78b99cb4cbac89fdfea70d.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "ce9be20c9c42ed83e5a2d5f0fd916477b0668fb92778fe2e7572f3623028baf2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a438db73301a7ace70884d5e86e645cf7dad88184f957b63665224cbdba4e56b", "generation_seconds_batch": 16.80922094907146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6217641830444336, "target_z": 1.1562498567676702, "ce": 6.300607204437256, "genuineness": 4.344170093536377, "identity": 0.6810393929481506, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ce9be20c9c42ed83e5a2d5f0fd916477b0668fb92778fe2e7572f3623028baf2.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "cebdca857df349bbdaf8fda2e8ddb478a5fa05463b6d6839424960356c9ed106", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "85230bc48823d12c995d9b2bf245c153e95f3676e8c15d7db817b449da177d31", "generation_seconds_batch": 22.97839677601587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4150304794311523, "target_z": 2.109375071616165, "ce": 4.800683498382568, "genuineness": 3.6475296020507812, "identity": 0.5666508674621582, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuando estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cebdca857df349bbdaf8fda2e8ddb478a5fa05463b6d6839424960356c9ed106.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "cec1d11875f6d8d975a283a61ae26753a34a87b8a380840e67ebf9ec9f59ec1b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a63debbc3c3ba168c6b9cc7da063eb428d82181395b155f57d7940f5ef91a479", "generation_seconds_batch": 21.594669436104596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5464661121368408, "target_z": -0.13574213826388667, "ce": 6.117925643920898, "genuineness": 4.432034492492676, "identity": 0.4491094648838043, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cec1d11875f6d8d975a283a61ae26753a34a87b8a380840e67ebf9ec9f59ec1b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "cee8482fc7c9876d7ee570caf7161310a0a5f9efacd19366cbe872041d523305", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bcddc319b1dff6238858716d7611bd133ffbee97bbb6c34b7db14432e41796bb", "generation_seconds_batch": 16.891833167988807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.496886968612671, "target_z": -0.19531246419191758, "ce": 6.109982967376709, "genuineness": 4.469690799713135, "identity": 0.6385350227355957, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cee8482fc7c9876d7ee570caf7161310a0a5f9efacd19366cbe872041d523305.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "cf56740850bfc1ca8de3ec682e6368e804ac7e64584463701d259f538caf1da3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "03e772859c587419548defbea7b09b5ebbe239642f4f4ca81eaffab9977435a4", "generation_seconds_batch": 22.448595992987975, "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": 5.879645347595215, "genuineness": 4.507347106933594, "identity": 0.8422902822494507, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cf56740850bfc1ca8de3ec682e6368e804ac7e64584463701d259f538caf1da3.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "5454b2632c36a5e1f985dabd6f579b4ba540e5807eecb3ba7b44d0bc93e49da2", "generation_seconds_batch": 16.646875888924114, "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": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.5000658631324768, "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_high/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "cf8d6da37071da7dfac91af3f590d1618ee4ed3c7fb442a6de56f779672faffc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3b73997fb36e14561803e0980c32be9427d9789dc88c60a9250c06be6946844f", "generation_seconds_batch": 24.822336797020398, "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": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.7518844604492188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cf8d6da37071da7dfac91af3f590d1618ee4ed3c7fb442a6de56f779672faffc.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "cff16aa1d2336c95c3889b6404cee8dd8e6b4e7f3c7c71d624d587242a1c9607", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "3afb19916e4acaf73177b8f87525f96dc19369f4103deb7a6cd2b4d7c79ae5ab", "generation_seconds_batch": 22.99340703105554, "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": 6.252951145172119, "genuineness": 4.017816066741943, "identity": 0.4188317358493805, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/cff16aa1d2336c95c3889b6404cee8dd8e6b4e7f3c7c71d624d587242a1c9607.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "d00a767ea94f1827317147631d5650479609ae753cfe9d8c4d51afd914e490ed", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e192095df0bcc5c1043fd0f53161c8769f7ae09d0bc0da40eb455600b271fa0a", "generation_seconds_batch": 16.69498343905434, "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": 6.054384231567383, "genuineness": 4.632867813110352, "identity": 0.7845937609672546, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d00a767ea94f1827317147631d5650479609ae753cfe9d8c4d51afd914e490ed.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "d0a804d4caa8d9680d6aedb6a807c71c07857af8a1f37f327092e37dc8693f77", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "505507bf8b0e8fc3b0d2a6b8cff4bc80aeebba38772a42eb1165d907a6712786", "generation_seconds_batch": 17.201562503003515, "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": 6.09409761428833, "genuineness": 4.6203155517578125, "identity": 0.6442174315452576, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d0a804d4caa8d9680d6aedb6a807c71c07857af8a1f37f327092e37dc8693f77.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d0f80a76392e0418e2d91728da6b17f1eb8f7d62def68b69d94cf39035956d43", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "38162f46b1289b190a98f43945e83269f5cca72b5d6997823350345eb9c820b0", "generation_seconds_batch": 17.705933527089655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8948559761047363, "target_z": 1.4843750716161648, "ce": 6.24500846862793, "genuineness": 4.406930446624756, "identity": 0.7321522831916809, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d0f80a76392e0418e2d91728da6b17f1eb8f7d62def68b69d94cf39035956d43.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "d0fc1ce3e0760b006a972bbd79cad85dc2cf401323bbd5fc93ad7918afff274c", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e739cfc742e0c76a77169197f924ea352e892c17084aeb2ec03888b027d802b1", "generation_seconds_batch": 17.168418818037026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.707395076751709, "target_z": 0.05761720988005152, "ce": 5.728734970092773, "genuineness": 4.325342178344727, "identity": 0.5260457396507263, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d0fc1ce3e0760b006a972bbd79cad85dc2cf401323bbd5fc93ad7918afff274c.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "d1127955e69f8ffb7b670024b33fc49063577e63d1ddc9c21a207a1dead4ab37", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ddb5c7b2ccac2247a82c1957b1a58f1a8f7f0a8a7be2e55c275305a29aa74942", "generation_seconds_batch": 21.237716658972204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50606107711792, "target_z": 2.2187501432323296, "ce": 5.49045467376709, "genuineness": 4.18099308013916, "identity": 0.857551097869873, "hypothesis": "Je trouvais la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d1127955e69f8ffb7b670024b33fc49063577e63d1ddc9c21a207a1dead4ab37.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "d11596dcc175b0815413825030ff292f9080624ba0ded87eacb9067ffb6414d7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8bb039945962f16547411d6fdac612e35b1df2eeb8321943b29d73b03ba6494b", "generation_seconds_batch": 18.021841699024662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.028440237045288, "target_z": 0.4433593481439382, "ce": 6.086154937744141, "genuineness": 4.15588903427124, "identity": 0.4994041919708252, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d11596dcc175b0815413825030ff292f9080624ba0ded87eacb9067ffb6414d7.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "ab5b517e72e852cf4c7bfccd10a985ad5b9f3bbf48b1d743980ecf5470ae6ebb", "generation_seconds_batch": 17.611844150000252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7648122310638428, "target_z": 1.3281249283838352, "ce": 6.3641486167907715, "genuineness": 3.848362922668457, "identity": 0.5264446139335632, "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_high/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "d1a4b5fc25c61ca9ce3d89bdba558dc03008c2627191700278b7da38432fb272", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4a8df8d80661a3e3404cde2df725331ab4eb43696956e1e264edfbf585e91a7e", "generation_seconds_batch": 17.269254012033343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6748842000961304, "target_z": 0.018554745688133947, "ce": 6.086154937744141, "genuineness": 4.6956281661987305, "identity": 0.8794388771057129, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d1a4b5fc25c61ca9ce3d89bdba558dc03008c2627191700278b7da38432fb272.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d1df152b85ef3243529376da5cae72ea77ab805ea001f26f2cc8fece3b58325e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e09a6699b500bd218a81d6fdd7b860affd803cf0d1c44659ca5d7c8f7aa4e458", "generation_seconds_batch": 16.436631727963686, "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": 5.8875885009765625, "genuineness": 4.306513786315918, "identity": 0.5773277282714844, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d1df152b85ef3243529376da5cae72ea77ab805ea001f26f2cc8fece3b58325e.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "5dde3debc5e7af326070837b082a3f9bd8528305ffb682588c85965827905fa4", "generation_seconds_batch": 17.715684288996272, "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": 6.3641486167907715, "genuineness": 4.444586753845215, "identity": 0.525115966796875, "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_high/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "d247dc4ea243acfb81d18afaec615ee1bdf8fe8e39fe7467bb47fdc1ea271ec2", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "68388eba8c63ea7acfedff222f63340bec7dc7c1c5b48ed2d99f4abff44b00d2", "generation_seconds_batch": 17.714737994014286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3649282455444336, "target_z": 0.8476563753282885, "ce": 6.149696350097656, "genuineness": 4.7458367347717285, "identity": 0.5999252796173096, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d247dc4ea243acfb81d18afaec615ee1bdf8fe8e39fe7467bb47fdc1ea271ec2.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "d254fb65fb6ff2aacad557404f0b4775f820474a797c2ab0e081fbb6806ea613", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ae018ebdcd7538ed67185425b7763026b9dfb6afbe76183999427d628b890b26", "generation_seconds_batch": 24.93322793301195, "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": 5.871703147888184, "genuineness": 4.20609712600708, "identity": 0.5209590196609497, "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_high/d254fb65fb6ff2aacad557404f0b4775f820474a797c2ab0e081fbb6806ea613.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "d257adcc84efbfbcd741d33bf258b07dc8099d34656b4dcc1c750002697f6b89", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b476eb969ceacbbf612849a7d4aa52007e436e92767a98527b057ab8b0f4dfb0", "generation_seconds_batch": 17.909289096016437, "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": 6.324435234069824, "genuineness": 4.457139015197754, "identity": 0.35062116384506226, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d257adcc84efbfbcd741d33bf258b07dc8099d34656b4dcc1c750002697f6b89.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "d2f0c348e5b61b5779203b2e6ea3a80b6757297d71c3a09f732f0c42f94ba769", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6e8a514d2c73a27ea4c0cec4bb0721dfd24c38721d69f8e9e485ca1d944b4446", "generation_seconds_batch": 24.377222332987003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7561614513397217, "target_z": 0.11621097778409273, "ce": 6.316492557525635, "genuineness": 4.400654315948486, "identity": 0.584219753742218, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d2f0c348e5b61b5779203b2e6ea3a80b6757297d71c3a09f732f0c42f94ba769.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "d360693e22469022b7fd9cd9e9079a683b671a80c530865a287ca4c2a7978405", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1867ec1a26e4251f63e7605d52b623d988c8d3e56d7d5c72d641c142b6eac126", "generation_seconds_batch": 16.75442807096988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892534255981445, "target_z": 1.1171875358080825, "ce": 6.284721851348877, "genuineness": 4.6705241203308105, "identity": 0.6011273264884949, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d360693e22469022b7fd9cd9e9079a683b671a80c530865a287ca4c2a7978405.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "d39767316117297e29731c984767e17de8ae7a154bb78020ff89e459bc2d8275", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "89d2c58637de8c61d67bfee59ed047af1e50e3a98c8a65c7c47152139a295792", "generation_seconds_batch": 24.96598624100443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5891366004943848, "target_z": -0.08447269877209788, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.7597018480300903, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d39767316117297e29731c984767e17de8ae7a154bb78020ff89e459bc2d8275.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d3ca1bfbe3bd9d77f49fab0771bc2e6df3716c7c9985a54884657a22ba0bb550", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8231e3820ef0a53cbba760111f188e423b71149fe143582d1c552a405a2bf579", "generation_seconds_batch": 21.577313233050518, "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": 6.070269584655762, "genuineness": 4.3127899169921875, "identity": 0.7318345904350281, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d3ca1bfbe3bd9d77f49fab0771bc2e6df3716c7c9985a54884657a22ba0bb550.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "1dbfd92b99f58fe85ebdcb9480f07cb51d5bdf58cccb20c779022bd9b336a820", "generation_seconds_batch": 15.875765564967878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1113431453704834, "target_z": 0.542968839520206, "ce": 6.300607204437256, "genuineness": 4.68307638168335, "identity": 0.4352073669433594, "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_high/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.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": "d46c79040a85a8c85b1f511b6e94347509de8a5e8ae5e5b1b7d88f7896526cb7", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "36605892373bd0c2173dd6d9710973e80ed773fee30c552eed92b9894d1457e7", "generation_seconds_batch": 17.993950573028997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868847131729126, "target_z": 1.4531249283838352, "ce": 6.316492557525635, "genuineness": 3.948779582977295, "identity": 0.5750565528869629, "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": 0.120000000000001, "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_high/d46c79040a85a8c85b1f511b6e94347509de8a5e8ae5e5b1b7d88f7896526cb7.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "d49050decf020a44d85d654e02454e0af3729f501ae77bb202e5c70009b1c8aa", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9d7e2861ef7bf5aa6c62824a26fca28ac59691ed817682930cadf4cebcee161e", "generation_seconds_batch": 17.168418818037026, "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": 6.038498878479004, "genuineness": 4.262581825256348, "identity": 0.7075996398925781, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d49050decf020a44d85d654e02454e0af3729f501ae77bb202e5c70009b1c8aa.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "d497b40bef5ab19b606a929b32695f56da8a1c74d3d3a2a086b26d22ddfbdfd2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f9e455afccecac2462ee3dc23ce043e87f6d404ae81d8aabfe78e90b0754143f", "generation_seconds_batch": 23.063927563955076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.988558292388916, "target_z": 4.0, "ce": 5.935244083404541, "genuineness": 3.5941832065582275, "identity": 0.24639280140399933, "hypothesis": "Ik heb een klein doosje gevonden. Het stond naast het oude raam. We kunnen samen in kijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d497b40bef5ab19b606a929b32695f56da8a1c74d3d3a2a086b26d22ddfbdfd2.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "d4d046c673f61b6608e2ac1738844a91fe4aacd13249132dac38a8a4f33a0416", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "63f52591dfd515d00a4feae2b498c0cb6d1dee2de5477f6ba6e212cf874ce233", "generation_seconds_batch": 24.00642042804975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5001380443572998, "target_z": -0.19140623209595878, "ce": 6.435632705688477, "genuineness": 4.432034492492676, "identity": 0.6521220207214355, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d4d046c673f61b6608e2ac1738844a91fe4aacd13249132dac38a8a4f33a0416.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "d4de78b9e6d5defb12bc690888253a5907ab0409517828ce56e99b07d186f822", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ebb3ce32f004350a4967aaa9840fef4a04cb875e7cba1629d253ce30364c003b", "generation_seconds_batch": 18.118048656033352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3974390029907227, "target_z": 0.8867186962878764, "ce": 6.205295085906982, "genuineness": 4.21237325668335, "identity": 0.3466787338256836, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d4de78b9e6d5defb12bc690888253a5907ab0409517828ce56e99b07d186f822.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "d599a6ddc0a9cdc649298f7921f11356bce839d179a11b4076b1bdb82d95cab2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b1d7a3a2480ac2e48247a5344730f041497d5daa6982c21e1df3c01087367eb6", "generation_seconds_batch": 23.283530917018652, "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": 6.062326908111572, "genuineness": 4.6956281661987305, "identity": 0.7007925510406494, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d599a6ddc0a9cdc649298f7921f11356bce839d179a11b4076b1bdb82d95cab2.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "d5abd0c15c85f7ad29c24a3431bf80baad02132810d06f52e78b1a9aca45bef6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4a1764fbd7e906dae82e01aa403d81bcc0a2ff3b8b4d8aeea355572aa91ae3df", "generation_seconds_batch": 18.226447381079197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8831875324249268, "target_z": 2.671875071616165, "ce": 6.24500846862793, "genuineness": 4.293961524963379, "identity": 0.4422244131565094, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d5abd0c15c85f7ad29c24a3431bf80baad02132810d06f52e78b1a9aca45bef6.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "d5cc09082db8cee557f3ac2a5d977edb8e27aee65a6275cdfa031c862760115f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "56946067cc0b14625e20cc91a860f1d78b71153ea1582f0a2809c79da48ad928", "generation_seconds_batch": 21.851496356073767, "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": 5.911416053771973, "genuineness": 4.532451152801514, "identity": 0.799771249294281, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d5cc09082db8cee557f3ac2a5d977edb8e27aee65a6275cdfa031c862760115f.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": "d5e568dcc5c5763079e33963dbbef053e8f92d732dbcbb477b13393bd1be96d3", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "24f6d98fc0145415eabacc00dc906bf5d6ad554841b54d45dadc9c366c6d68c6", "generation_seconds_batch": 23.385942174936645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 6.268836498260498, "genuineness": 3.8169827461242676, "identity": 0.6058895587921143, "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": 0.120000000000001, "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_high/d5e568dcc5c5763079e33963dbbef053e8f92d732dbcbb477b13393bd1be96d3.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": "d6394dee08e3e42281d8529ee3015ab2337c96a30aa6a1cd7beca017a46ed038", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e36f3fd5b03d3dcde5dc9b42a0f7711399418b196c31dea1e59db2b313bb200", "generation_seconds_batch": 17.142593416036107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.31941294670105, "target_z": 0.792968839520206, "ce": 6.221180438995361, "genuineness": 4.68307638168335, "identity": 0.5568673014640808, "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_high/d6394dee08e3e42281d8529ee3015ab2337c96a30aa6a1cd7beca017a46ed038.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "d6494925bcad4cc1c67f164ef55766b36f860315bfdd1430a53db4fb14672e84", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "056f7300da48abf3ecdb27635de1336636d23bce715eea1b4453f0787ca8798a", "generation_seconds_batch": 20.694722806103528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5204010009765625, "target_z": 3.4374997135353405, "ce": 5.77639102935791, "genuineness": 3.5345609188079834, "identity": 0.46714839339256287, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d6494925bcad4cc1c67f164ef55766b36f860315bfdd1430a53db4fb14672e84.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "d6511e3e0f44fd5184fd3072f014c11fd3d2b1f4c1892abe6d093c69ad115da3", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a103f2ae285d410105a86f2915fa8f482d45c6f91f06f7211f632208b5444572", "generation_seconds_batch": 16.628088337020017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5177292823791504, "target_z": 1.0312498567676702, "ce": 6.157639026641846, "genuineness": 4.595211505889893, "identity": 0.45631739497184753, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d6511e3e0f44fd5184fd3072f014c11fd3d2b1f4c1892abe6d093c69ad115da3.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d72a4d0753e6e9a9f44b9ce2c47c00038690c1f95514560576fbba6a9c33361e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "cc2c1f2bcecd8a088eef79b281e63d9623677bedb478fbfd32a53a49c6c4eb4a", "generation_seconds_batch": 23.94466196396388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.702462196350098, "target_z": 3.6562498567676704, "ce": 5.526196479797363, "genuineness": 3.7416701316833496, "identity": 0.48471081256866455, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d72a4d0753e6e9a9f44b9ce2c47c00038690c1f95514560576fbba6a9c33361e.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "d7370d503e565446778ce89115fed2efe9488bf59472a40bb3692065ae9e9363", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c2aff584108fbafb7aeb75376abf9f87f5d2962b1a129483d99aa6c54b741eaf", "generation_seconds_batch": 23.46593542804476, "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": 6.141753673553467, "genuineness": 4.105680465698242, "identity": 0.5218114256858826, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d7370d503e565446778ce89115fed2efe9488bf59472a40bb3692065ae9e9363.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "d88411ec9ae42ab2a096452a82ea5876694a87fa7c6afe232e6446ce0aeee17f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2c186b22b5dfa351bc55ebc90a4a643776a01e8a123c8c40867c395752d28fc1", "generation_seconds_batch": 23.961544070974924, "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": 6.149696350097656, "genuineness": 4.482243061065674, "identity": 0.6412631273269653, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d88411ec9ae42ab2a096452a82ea5876694a87fa7c6afe232e6446ce0aeee17f.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d8a3b972ca646aed2bf2aff3e8c03e52c47d7c9a6174d2721f123d89bbcb639f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f26437a697bc1345a6d1f4f2d6d60fd1996fd70d892c517d835eb5fb49c8eda1", "generation_seconds_batch": 16.698649651021697, "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": 6.086154937744141, "genuineness": 4.300237655639648, "identity": 0.5569963455200195, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d8a3b972ca646aed2bf2aff3e8c03e52c47d7c9a6174d2721f123d89bbcb639f.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "d8c3dd941048d98534250340f7489d40477c90df13b5985e22583ff5b4973450", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "421b3f39b15ad272b8ec2bb5b1347658cea46298f12d0fcfe8315bb303a6d514", "generation_seconds_batch": 17.00179603707511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.078212261199951, "genuineness": 4.18726921081543, "identity": 0.6605983376502991, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d8c3dd941048d98534250340f7489d40477c90df13b5985e22583ff5b4973450.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "d8d86a23e814425a6ac6ae2b95c64dca30389e7b13393801cdb121f0b1454515", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9f2f6e6e0c4d00e3a1a32a1699aca146b5dc1e1b3d820001c6277d563b0cadfd", "generation_seconds_batch": 17.9043094139779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.452707529067993, "target_z": 0.9531249283838351, "ce": 6.1894097328186035, "genuineness": 4.15588903427124, "identity": 0.23581525683403015, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d8d86a23e814425a6ac6ae2b95c64dca30389e7b13393801cdb121f0b1454515.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": "d8f2ab76e579ad116ca7c4ec9d3ed7f876e3b20e9c4abe1c7b6a0795ca3d8ce2", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a9a3dc3450388418c751c1269a1b35acc7aff47c806947a02e3fc139de3a383", "generation_seconds_batch": 18.935195220052265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8038253784179688, "target_z": 1.375, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.7184643745422363, "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_high/d8f2ab76e579ad116ca7c4ec9d3ed7f876e3b20e9c4abe1c7b6a0795ca3d8ce2.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "d90a6f83971a790a496bc3b74aa1f13e9e32d5feb5534417a5931dc444c4c935", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fcc8d6f1289c4590d45f7afa961e441c72031d3755520ff1630365c8b66bd9d9", "generation_seconds_batch": 23.633638267987408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2088756561279297, "target_z": 0.6601560888636291, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.5554137229919434, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d90a6f83971a790a496bc3b74aa1f13e9e32d5feb5534417a5931dc444c4c935.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "d942749df45d3e6adc0f4f4d9e53361be502f141ac9f713bce8ca589252f00e4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2906b793fb4be127b4c09d744dbfe4d3e6c796cff9229643efbf687e9284dffa", "generation_seconds_batch": 22.42010622296948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.039239883422852, "target_z": 2.859375071616165, "ce": 5.502368927001953, "genuineness": 3.7165660858154297, "identity": 0.26439106464385986, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d942749df45d3e6adc0f4f4d9e53361be502f141ac9f713bce8ca589252f00e4.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "d963438b49e4a4428cf3bd8c2f9d0a6d4e558fcb157d44421826181ade8f066f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e00faaaff35a3581d3e0c9402e09e14fc397f6e49cf06f8fb0cd89ab4426fb5d", "generation_seconds_batch": 22.310627399012446, "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": 6.2767791748046875, "genuineness": 4.400654315948486, "identity": 0.5593155026435852, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d963438b49e4a4428cf3bd8c2f9d0a6d4e558fcb157d44421826181ade8f066f.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "d9d55e940c093e0d4e454c204be4bcfa50606c5bf990f644f1a7dc02b01b92e2", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a0f743190eef65f2050dd2de0b22b558db6186bbf044358fa83e4a580473d3f3", "generation_seconds_batch": 18.021813014987856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6521265506744385, "target_z": -0.008789022215907268, "ce": 6.109982967376709, "genuineness": 4.262581825256348, "identity": 0.5518648028373718, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d9d55e940c093e0d4e454c204be4bcfa50606c5bf990f644f1a7dc02b01b92e2.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "d9eb7d8f92587414c7c10316532fa85b722e72b1b6b44d4ef6041e71b6993171", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6fb282a178ccf91bd85798ed2afcce221a34c4c94aacc7ccff75333388f89d83", "generation_seconds_batch": 16.69498343905434, "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": 5.935244083404541, "genuineness": 4.444586753845215, "identity": 0.6415035724639893, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/d9eb7d8f92587414c7c10316532fa85b722e72b1b6b44d4ef6041e71b6993171.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "da158e13e41716e6de7881d8486d9d8ed0b438ec27886648828a288c8be6ca5a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ded5f1e11b3be7695f8762d1635ecfc065132ac5d0f221165bfb98519753dd51", "generation_seconds_batch": 23.20532251300756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3760173320770264, "target_z": 2.0625, "ce": 5.903473377227783, "genuineness": 4.494794845581055, "identity": 0.7706374526023865, "hypothesis": "La porte t'est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/da158e13e41716e6de7881d8486d9d8ed0b438ec27886648828a288c8be6ca5a.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "da1aff2b2231392714bda7d38f739d74e5bda2fb8bb692b5e6d70f0bf49bea73", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9ead7905acf4fd4008c9a812d7cda6ebc35247799a3f9f75564ac523a5b879cd", "generation_seconds_batch": 17.32587613002397, "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": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.5850539803504944, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/da1aff2b2231392714bda7d38f739d74e5bda2fb8bb692b5e6d70f0bf49bea73.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "da5a171c6491faf3004dbbdf0f8e15c380d61b2576bdbd21255c7c66a91ce559", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "13d6030be32abbcfec4add567f5ee815e3ff86a2536655912d33d8daec8b9516", "generation_seconds_batch": 23.056112954975106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8219959735870361, "target_z": 0.19531246419191758, "ce": 6.308549880981445, "genuineness": 4.268857479095459, "identity": 0.33856239914894104, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/da5a171c6491faf3004dbbdf0f8e15c380d61b2576bdbd21255c7c66a91ce559.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "db181600ad516cd1873a0902f1fb5244da5246077a9e5233031857cf2ee3dacd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d95c49f51f1528d03966c3659e87ffa4f7cd9214db568f2e0f1f96209be04141", "generation_seconds_batch": 25.435089610051364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5883238315582275, "target_z": -0.08544925679608759, "ce": 6.054384231567383, "genuineness": 4.18099308013916, "identity": 0.7545232772827148, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/db181600ad516cd1873a0902f1fb5244da5246077a9e5233031857cf2ee3dacd.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "c34093ae731d3bbe70ea400d92e5923ea40ccbfaf9b55d8a96eba7c52a25305e", "generation_seconds_batch": 18.48441807902418, "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": 6.427689552307129, "genuineness": 4.168440818786621, "identity": 0.3885596990585327, "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_high/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "db5cd1e23349b18bc976ea150f0fa27f92d46b81bb3b45d53482ae07caf7d534", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "567a4843e958762329a521ac39a065a5a0e2ebb7a2d5c590c5ac8b210b421dd5", "generation_seconds_batch": 17.701858429936692, "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": 6.332377910614014, "genuineness": 4.632867813110352, "identity": 0.6254293918609619, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/db5cd1e23349b18bc976ea150f0fa27f92d46b81bb3b45d53482ae07caf7d534.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "db74fa3ac1f8197431a56cc2676c46f7e45d04f574e07ea7fc72f1935e402447", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "08af2c780d136da98a1fa031997d02124fed6453fedb7b82eb176c876aa1d033", "generation_seconds_batch": 21.598848772002384, "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": 6.157639026641846, "genuineness": 4.595211505889893, "identity": 0.6303427815437317, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/db74fa3ac1f8197431a56cc2676c46f7e45d04f574e07ea7fc72f1935e402447.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "db7f72fcfb5c3a6e4be5a487b55eafbf5c69a63b1837991b42cbe91b043022ee", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6a4b0462d477ec14a9af462f9b5bc8d7770374cf39dd31fd3ad72ec29a7dfcf8", "generation_seconds_batch": 17.492815589997917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3291661739349365, "target_z": 0.8046875358080824, "ce": 5.911416053771973, "genuineness": 4.545003414154053, "identity": 0.75935959815979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/db7f72fcfb5c3a6e4be5a487b55eafbf5c69a63b1837991b42cbe91b043022ee.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "db91c845a73401ba8fe714096f6300b599eff346c41a1acd6dfc8e87715d6973", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fdd6fd702ec809cc12bce32b7d276c23a7516c003f0555fd822c8394333cbec3", "generation_seconds_batch": 17.208584030973725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.202373504638672, "target_z": 0.6523436246717115, "ce": 6.038498878479004, "genuineness": 4.268857479095459, "identity": 0.4722222685813904, "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_high/db91c845a73401ba8fe714096f6300b599eff346c41a1acd6dfc8e87715d6973.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "dbaa41f88f3527c6140bbe8aec7e27cd99da1fb6dcac5c5a180f92f21a3e3202", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "5d57341dce823afbf76158e6d33a123d0fd8c78291b9fe3e5143a01d4af4aff8", "generation_seconds_batch": 24.58434047398623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5437381267547607, "target_z": 1.0625, "ce": 6.1894097328186035, "genuineness": 4.400654315948486, "identity": 0.649606466293335, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dbaa41f88f3527c6140bbe8aec7e27cd99da1fb6dcac5c5a180f92f21a3e3202.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "dbb05caa7440c6d5c5ed92ca9e4b1d525b7d0b018c603ae47c0f38de619d27e6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "51972e9f9339fbee8a934fed95f24acf8379c564bf10923abc504de29fdb3359", "generation_seconds_batch": 22.718986788997427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.884523391723633, "target_z": 3.875, "ce": 5.3514580726623535, "genuineness": 3.6663577556610107, "identity": 0.4310552477836609, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dbb05caa7440c6d5c5ed92ca9e4b1d525b7d0b018c603ae47c0f38de619d27e6.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "dc6288329d5a8db6b95158ced14c171d49b5b2602b7c04494c515024e4080b4a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9ed221028f886fb673102e7ae09584d4a6c3b75a40bb7345c48bad6560b5b6b5", "generation_seconds_batch": 17.108321174047887, "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": 5.927301406860352, "genuineness": 4.381826400756836, "identity": 0.6764615774154663, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dc6288329d5a8db6b95158ced14c171d49b5b2602b7c04494c515024e4080b4a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "dc73b90f618c4571f99d106dd50b6381eac6180307b79fd190d9a0bcc546212a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9734f7031ed671c4d7453c7deb743353b20179f4495d204ad1768291a186084a", "generation_seconds_batch": 24.001239183009602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9585418701171875, "target_z": 0.35937507161616483, "ce": 6.332377910614014, "genuineness": 4.469690799713135, "identity": 0.5620017051696777, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dc73b90f618c4571f99d106dd50b6381eac6180307b79fd190d9a0bcc546212a.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "dc89946edad5ceddeb2a126c9f3c9010819958152835e2e0239159580ffbfd30", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "00a75f2813d28fb1acd62b1ff57c46d69c5d57bc780f3adb1161121c501b1f5d", "generation_seconds_batch": 20.80498196708504, "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": 6.300607204437256, "genuineness": 4.607763767242432, "identity": 0.28867220878601074, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dc89946edad5ceddeb2a126c9f3c9010819958152835e2e0239159580ffbfd30.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "dc96ed6c5e9ac5bfa6e53dd094c511d299d44486576e8c650ea4be1ebf5ca64a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "82e4ef8eb78c5b986e5088cf2803b05eacd63176b2deb7d17f126a8452735b03", "generation_seconds_batch": 17.71516993304249, "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": 6.022613525390625, "genuineness": 4.507347106933594, "identity": 0.8060072660446167, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dc96ed6c5e9ac5bfa6e53dd094c511d299d44486576e8c650ea4be1ebf5ca64a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "dc9c58708a24e3d6051a1dc9fbc6086909405bc7288cb9679ab625b55880bd2d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "79d3a3fd974097531ee2776ad434042c9130ee895d43030df99341601089c398", "generation_seconds_batch": 25.081651952001266, "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": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.6133825778961182, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dc9c58708a24e3d6051a1dc9fbc6086909405bc7288cb9679ab625b55880bd2d.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "dca2926149e439d7314cbe1cb6f2ddc91989e61fcf23c395a9b57d83629f9505", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "84a551852a0d2d3bca63450419612776fd3d80e2813c0a708b2e07499a721d2c", "generation_seconds_batch": 23.64414112304803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6607773303985596, "target_z": 1.2031249283838352, "ce": 5.007937431335449, "genuineness": 4.118232727050781, "identity": 0.4514312744140625, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dca2926149e439d7314cbe1cb6f2ddc91989e61fcf23c395a9b57d83629f9505.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "dcafb05cbd0da1741ee03165582e4e484c0f20629cfb92485789c642a20cb630", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8d5ff5384faae03fb428cbe00b458c815d32a159d19b9271cd6a47d962058d53", "generation_seconds_batch": 17.688341216999106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6217641830444336, "target_z": 1.1562498567676702, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.8437870144844055, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dcafb05cbd0da1741ee03165582e4e484c0f20629cfb92485789c642a20cb630.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "dcd54c293be397581909c8984cf32551c1570eb3fa20b192028a1359fa463dd1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "43392a352c1f897e0d182834e9f4c115561912ca4d3d0b7c7085bfc6d4fc94fc", "generation_seconds_batch": 16.91786885797046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8103275299072266, "target_z": 1.3828124641919175, "ce": 6.300607204437256, "genuineness": 4.331617832183838, "identity": 0.5218799114227295, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dcd54c293be397581909c8984cf32551c1570eb3fa20b192028a1359fa463dd1.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "dcec6bee77f3098d5ca5a673fa1e0be07472b9c4efa9f10a26114cf076fba040", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "889b8069e8051ea2bd89329470a781f3951b26855dcab0b4583a9cfe05e32191", "generation_seconds_batch": 22.414218489080667, "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": 6.284721851348877, "genuineness": 4.808597087860107, "identity": 0.6870388984680176, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dcec6bee77f3098d5ca5a673fa1e0be07472b9c4efa9f10a26114cf076fba040.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "dd0bb485487267cd4a7d435556e2065dc79dafe88099ac96765308e8b034046f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d27a206a015f677e0ddeb55ccbe22c8faf23c11bd91df8a479eae73bff568759", "generation_seconds_batch": 23.318785760900937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.245973587036133, "target_z": 1.9062498567676702, "ce": 6.054384231567383, "genuineness": 4.2249250411987305, "identity": 0.722141683101654, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dd0bb485487267cd4a7d435556e2065dc79dafe88099ac96765308e8b034046f.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "dd9a9f27eb201d23e40710a28bba16849d929953715473d026cbfcabcdbb4d01", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7061c7f2eafee3cd4c31a9fce38fbd4cd2d18a18af63ce9e813cc189c861b26e", "generation_seconds_batch": 23.17735840496607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8423153162002563, "target_z": 0.2197265580239897, "ce": 6.1894097328186035, "genuineness": 4.7332844734191895, "identity": 0.7675464153289795, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dd9a9f27eb201d23e40710a28bba16849d929953715473d026cbfcabcdbb4d01.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "ddb595eb62719865cea1e16d73835b9a08a1a75681d7a8da1bb6d90ee61b47ec", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "566e37dffc1bce17bb29ba048c233209e9b38621958cdf1c9f844552a87b799b", "generation_seconds_batch": 17.257350855972618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.024899482727051, "target_z": 1.6406249283838352, "ce": 5.895530700683594, "genuineness": 4.243753433227539, "identity": 0.6933777332305908, "hypothesis": "He encontrado en la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la vida.", "wer": 0.23529411764705882, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ddb595eb62719865cea1e16d73835b9a08a1a75681d7a8da1bb6d90ee61b47ec.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "ddd2c534cc83f89cf6c733323e3c907c36406bfd2350786c7641a877d1c7ed32", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "efe6c1161ff71d5b6c6bf54676bd12e6ac352f0a4a2a6f163dc73e53ba387d24", "generation_seconds_batch": 15.3653919310309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.60225772857666, "target_z": 1.1328124641919175, "ce": 6.252951145172119, "genuineness": 4.068024158477783, "identity": 0.7498178482055664, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ddd2c534cc83f89cf6c733323e3c907c36406bfd2350786c7641a877d1c7ed32.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": "de79e5287e80fa3ac5a9313aad91876f0cafa5d96d2a1e2c35e76923fe1f0765", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "586be853aa274ded8593b658e1e291e439f1cf93d602745ada963a74ff92041e", "generation_seconds_batch": 20.593635026016273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4150304794311523, "target_z": 2.109375071616165, "ce": 6.165581703186035, "genuineness": 4.20609712600708, "identity": 0.5104252099990845, "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_high/de79e5287e80fa3ac5a9313aad91876f0cafa5d96d2a1e2c35e76923fe1f0765.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": "de9e8ede2544dc980e9ad68159e34ac3c7d9198a93a551fad46702b8ddcbab49", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6191cb75e92e0e504b839f4353bbdc666e49aec2702702e1c61ebdb8f78749b4", "generation_seconds_batch": 22.96548067999538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0834193229675293, "target_z": 1.710937679040412, "ce": 5.935244083404541, "genuineness": 4.287685871124268, "identity": 0.4464086890220642, "hypothesis": "Las llaves están en la mesa de la cocina. Miren, 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_high/de9e8ede2544dc980e9ad68159e34ac3c7d9198a93a551fad46702b8ddcbab49.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "df507d104d33ac3beb542d935b3b91a2026705224d2d24ca27f57c4bf1d5af2d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "7f4f7faa00314d776128034651c3c16f63c584c91b2cf1268821b423cd99120c", "generation_seconds_batch": 21.646045641973615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.284986734390259, "target_z": 1.9531249283838352, "ce": 5.4944257736206055, "genuineness": 4.18099308013916, "identity": 0.3490264415740967, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/df507d104d33ac3beb542d935b3b91a2026705224d2d24ca27f57c4bf1d5af2d.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": "df5c11ff973e27d63622080e3e9bdb0316572e3e7c17f4b91789ca8220960aa3", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d908dd31189d2edc431664bbcf1fd0c35c2dbad3315561d28b01c5c3b8fab2f7", "generation_seconds_batch": 20.593635026016273, "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": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.5096247792243958, "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_high/df5c11ff973e27d63622080e3e9bdb0316572e3e7c17f4b91789ca8220960aa3.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "df736b46ccda618973b7e59496306dadecc60665f6ec90808f0aa860f84559e6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e794d6dae36b2701434ee1de29f0a5e8d5fd019e1f02cd21ebba2f4f30653296", "generation_seconds_batch": 24.205222241929732, "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": 6.181467056274414, "genuineness": 4.457139015197754, "identity": 0.500336766242981, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/df736b46ccda618973b7e59496306dadecc60665f6ec90808f0aa860f84559e6.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "dfa3d3f725d3674827ff9cd780c586e4c3373fc17d3d9f6fa12ea31cc49eeba7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2fbb87c7da17dc8facb0ec8ec410ff039cb021ef8a230d1bac9cce7ea8287bb3", "generation_seconds_batch": 17.234165358007886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0379037857055664, "target_z": 1.6562498567676702, "ce": 6.054384231567383, "genuineness": 4.287685871124268, "identity": 0.6426317691802979, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/dfa3d3f725d3674827ff9cd780c586e4c3373fc17d3d9f6fa12ea31cc49eeba7.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "e0170ab8c7efbf9c63fa66891a91cdfdd3445c2ce44c3239a587936bee2399c2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ff89b985ef19e18ce83b8291485fc032db4228fe9985058fd7c67846e3bfc89b", "generation_seconds_batch": 16.80922094907146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.901358127593994, "target_z": 1.4921875358080825, "ce": 6.395919322967529, "genuineness": 4.337893962860107, "identity": 0.43279916048049927, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e0170ab8c7efbf9c63fa66891a91cdfdd3445c2ce44c3239a587936bee2399c2.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "e0cad2c57758ba649fcd2a71c53e3c96f299d1609f90af19905306716e5078bb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7d73207468657b55f41bc1b4d396acb72d843cb41f70729849d6864e39c6d970", "generation_seconds_batch": 23.056112954975106, "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": 6.38797664642334, "genuineness": 4.432034492492676, "identity": 0.5769590139389038, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e0cad2c57758ba649fcd2a71c53e3c96f299d1609f90af19905306716e5078bb.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "e110e1bac97a0c4a47a1060e136ea6986f34484d131f3ff3dc51446eb15f00d7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5ce23bc3db35aa136998c66f24cfa6df1c538a29bdf037950d46fc680ed93ea2", "generation_seconds_batch": 25.912419773056172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680284023284912, "target_z": 1.2265626074242473, "ce": 6.348263263702393, "genuineness": 4.381826400756836, "identity": 0.612013578414917, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e110e1bac97a0c4a47a1060e136ea6986f34484d131f3ff3dc51446eb15f00d7.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "e17bea821544248e6651c608d14881b4c8afb72851c61847781c2df81764160f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "88be8985daa65d45299aab1994e596dffa95a07ece9f4986e808ad21ed90fedb", "generation_seconds_batch": 22.700801557977684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5147678852081299, "target_z": -0.17382818766414426, "ce": 6.403861999511719, "genuineness": 4.300237655639648, "identity": 0.657783567905426, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e17bea821544248e6651c608d14881b4c8afb72851c61847781c2df81764160f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "e1eb1e241b9d09d7892e2ffb769cefc222290b68c1bf2c68321ad8cdeea833a5", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "72d75af755f5c58a77caaf58853071d548f0420a52c4e27207640b124f59255c", "generation_seconds_batch": 18.119154921965674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9273667335510254, "target_z": 1.5234373925757527, "ce": 5.95112943649292, "genuineness": 4.319066047668457, "identity": 0.6885296106338501, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e1eb1e241b9d09d7892e2ffb769cefc222290b68c1bf2c68321ad8cdeea833a5.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "e2686534055c28d4d8f34287b290df7be780359576e8f63ec0e1f7a7a0c50aad", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1c6aedee08884d75f764141cb93df1e5716a9adf6901d8a352872d443f000813", "generation_seconds_batch": 22.718986788997427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.793492794036865, "target_z": 3.765624928383835, "ce": 5.111192226409912, "genuineness": 3.653805732727051, "identity": 0.43040260672569275, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e2686534055c28d4d8f34287b290df7be780359576e8f63ec0e1f7a7a0c50aad.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "e33aa49a1ea3ca1e13fbd0a2649c991b8a46185027f4f5f27fe4d2c516695435", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "a847c26f2ea31a34dce6e2e3034fdf826a8dd963bc04761d2547eb5db5f58655", "generation_seconds_batch": 24.513826826005243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.719296932220459, "target_z": 1.2734373925757527, "ce": 6.1020402908325195, "genuineness": 4.557555198669434, "identity": 0.5202860236167908, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e33aa49a1ea3ca1e13fbd0a2649c991b8a46185027f4f5f27fe4d2c516695435.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "e384e78d6e9a2de4100d9001624fb94bcb054fc701413173beb2e76603d8ab2a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "089a35a32329c320269c994ea6a090a1e303567b323a930bca0dfd6bca21437f", "generation_seconds_batch": 22.58199268206954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2608933448791504, "target_z": 0.7226563753282885, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.4111493229866028, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e384e78d6e9a2de4100d9001624fb94bcb054fc701413173beb2e76603d8ab2a.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": "e4771001d4c1639ecb481ad3559e49b221fa22a9243db78108fd284df5489cff", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f9025429bae11980d53a1af6666b098b1580a601fcb50ad885a9e71029edf65", "generation_seconds_batch": 23.312467143987305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9143624305725098, "target_z": 1.5078124641919175, "ce": 6.006728172302246, "genuineness": 4.306513786315918, "identity": 0.760157585144043, "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_high/e4771001d4c1639ecb481ad3559e49b221fa22a9243db78108fd284df5489cff.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "e4a8f83385bbfc40a5d4855cf320e88fc2aa2e88295599ea5ce84f1ce72e6d18", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8afc43042659d94998c2e36b2088db83fa44f4a0495e2ddafa25b2ffe442f389", "generation_seconds_batch": 16.58229361602571, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0414445400238037, "target_z": 0.45898427652777335, "ce": 6.229123115539551, "genuineness": 4.68307638168335, "identity": 0.6763903498649597, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e4a8f83385bbfc40a5d4855cf320e88fc2aa2e88295599ea5ce84f1ce72e6d18.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "e4e3ed12e65595d0c58c66c0d5087ab12ed6a64fbe0d4b2e89a3874500408388", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bfaf17452ebedd4502409a08ed080273499266b66d030fd570e71962a37dff6f", "generation_seconds_batch": 23.388810685952194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.901358127593994, "target_z": 1.4921875358080825, "ce": 5.585766792297363, "genuineness": 4.356721878051758, "identity": 0.5242306590080261, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e4e3ed12e65595d0c58c66c0d5087ab12ed6a64fbe0d4b2e89a3874500408388.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "e5131b6f9af45995c61f484e2302fd464370b2812579faf803d8b138503e7f8a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d88610e32d756e7c04cd8fbe874b795cc1b09c66f038e4023559deafca69cc56", "generation_seconds_batch": 23.61446019797586, "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": 6.268836498260498, "genuineness": 4.65797233581543, "identity": 0.6568364500999451, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e5131b6f9af45995c61f484e2302fd464370b2812579faf803d8b138503e7f8a.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "e554147ed45a35587564a7dc8039730ed53c57b44c87756f8c32ca431462baf1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d1161773025cdedc5fab20be981756deeda8fe4ec7ae40de5072a6c5c8bd08bb", "generation_seconds_batch": 22.310627399012446, "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": 6.332377910614014, "genuineness": 4.720732688903809, "identity": 0.31071728467941284, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e554147ed45a35587564a7dc8039730ed53c57b44c87756f8c32ca431462baf1.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "e55c4d67272b91937d4332a4aea935fda3a71b0e03e1997d1def8215a9856139", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e5b45595847d160e86e94f2f0aed2a6f15e0790db7c8f099fa900a36f62095a3", "generation_seconds_batch": 22.608567847986706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6542751789093018, "target_z": 1.1953124641919175, "ce": 6.125868320465088, "genuineness": 4.545003414154053, "identity": 0.6901214718818665, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e55c4d67272b91937d4332a4aea935fda3a71b0e03e1997d1def8215a9856139.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "e5962c6bc0f6421a953a66ebde715387c5dbbe92ecd69562da4fd4c566f10a98", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7971462b96f815aaff5679061272bc1927ed19e32834a6de47ce9df11d596407", "generation_seconds_batch": 24.165765804005787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7492527961730957, "target_z": 0.1079100913478506, "ce": 5.974957466125488, "genuineness": 4.6956281661987305, "identity": 0.6398946046829224, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e5962c6bc0f6421a953a66ebde715387c5dbbe92ecd69562da4fd4c566f10a98.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "e5ddf864bbde25b14c1d14f7c6023923a5ae2fd131baa82235dd0b97487fffbd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "629c3d86b46676902dc09fa617f16b62bb778231e3ff1eaa6b318697a3234724", "generation_seconds_batch": 17.975838333950378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2576422691345215, "target_z": 0.7187501432323297, "ce": 6.221180438995361, "genuineness": 4.833701133728027, "identity": 0.5337643623352051, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e5ddf864bbde25b14c1d14f7c6023923a5ae2fd131baa82235dd0b97487fffbd.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": "e63bdda04d3dc384beb3de2defaafc1dafd2b1e9f437c463bc0d6a96178563c9", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "614fd27470571089ce8ae193596f5c764774c10c800ac809177324d1ad06e860", "generation_seconds_batch": 18.848783902009018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.68678617477417, "target_z": 1.2343750716161648, "ce": 6.149696350097656, "genuineness": 4.193545341491699, "identity": 0.42490634322166443, "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_high/e63bdda04d3dc384beb3de2defaafc1dafd2b1e9f437c463bc0d6a96178563c9.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "dd6f4896326b4904937af6731028a6bbcdd9adf9ba08352d148bc69ac7bd3c8d", "generation_seconds_batch": 23.290208989055827, "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": 6.316492557525635, "genuineness": 4.118232727050781, "identity": 0.4844311773777008, "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_high/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "e6f9015bfe909ae01ae1e214c3bef1f78f36c2a3339c4ff6327d679d908ea2cb", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "78d5c979adb53e97864a886dea249409a6cd37889da64ae77c80b2d1fa639170", "generation_seconds_batch": 24.23062640801072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1484408378601074, "target_z": 1.789062320959588, "ce": 6.1020402908325195, "genuineness": 4.306513786315918, "identity": 0.6269201040267944, "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_high/e6f9015bfe909ae01ae1e214c3bef1f78f36c2a3339c4ff6327d679d908ea2cb.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "e7a366295092531c3b5200b7e4b459499311ef18a824d1e10a4ba3341ab6ea43", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "91dd04ce59cfb2baab1c784421e524a691577396bd151c76e5bc81545b473d3e", "generation_seconds_batch": 23.318785760900937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.862344980239868, "target_z": 1.4453124641919175, "ce": 6.308549880981445, "genuineness": 4.419482707977295, "identity": 0.3942381739616394, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e7a366295092531c3b5200b7e4b459499311ef18a824d1e10a4ba3341ab6ea43.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "e8cebeb74277cb633d9854fe59a936f29e8339f2455c400af828d2692aff4405", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ac618da4fa86cc920f65fc8aa52bce924d77c8c3ab0af05d5547635899c36a12", "generation_seconds_batch": 22.159214230021462, "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": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.4203318953514099, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e8cebeb74277cb633d9854fe59a936f29e8339f2455c400af828d2692aff4405.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "e8d773196d03d6bcd37545b16f2944d8bb81e15c79500eab3209c7dde544b659", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e71894ccc0ed5fe2a8f5b33e1156bb804b6f9bfae4771966cb47b39e45db752d", "generation_seconds_batch": 18.611502249957994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.309659481048584, "target_z": 0.7812498567676703, "ce": 6.117925643920898, "genuineness": 4.444586753845215, "identity": 0.28280553221702576, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e8d773196d03d6bcd37545b16f2944d8bb81e15c79500eab3209c7dde544b659.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "e8dadfe8a02ccd4ecd2f01337d6d48ad476d27736d977d61270d18fda6e571bd", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5ac7401965fb042e91828b7144e463f83d41807b1b83ace95281b5e151c52866", "generation_seconds_batch": 22.256560069974512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.751807928085327, "target_z": 1.3125, "ce": 6.332377910614014, "genuineness": 4.68307638168335, "identity": 0.6059541702270508, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e8dadfe8a02ccd4ecd2f01337d6d48ad476d27736d977d61270d18fda6e571bd.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "e93cd7e2cd01f785a86e9512462994d330d0e5a439bf780f849f7272f8690f69", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "794c2a40a5dd35bc2cf5bed87ac70375327d4460a91074c7ba992a0dbcb3ae67", "generation_seconds_batch": 23.202552985982038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7791526317596436, "target_z": 2.546875071616165, "ce": 5.990842819213867, "genuineness": 3.7667744159698486, "identity": 0.5235046744346619, "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_high/e93cd7e2cd01f785a86e9512462994d330d0e5a439bf780f849f7272f8690f69.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "e966d0c02f303eea0c68691fa8df27da74aff984b22affee2343411d24d3acc0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "be4c4fc346cbe7beaa7d7a014c686fadba8530dd877a5ccc3b434d948c4323dd", "generation_seconds_batch": 22.871165193035267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.350008487701416, "target_z": 2.0312498567676704, "ce": 5.276002407073975, "genuineness": 3.8044304847717285, "identity": 0.7131296992301941, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/e966d0c02f303eea0c68691fa8df27da74aff984b22affee2343411d24d3acc0.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "e9c28c53086265306ce94190510bd7fac152adb378fcb6f5092fb284d8beb14b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "df89b308d7155500497e67d342b560427bcc08601e126b0eaebd0a39f50d2abd", "generation_seconds_batch": 25.53896590997465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.245973587036133, "target_z": 1.9062498567676702, "ce": 5.708878040313721, "genuineness": 4.137060642242432, "identity": 0.35884857177734375, "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_high/e9c28c53086265306ce94190510bd7fac152adb378fcb6f5092fb284d8beb14b.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "8785651e6faeb7c353ec9aa85e20e72299cc966bbe525ed7f803c0638c1fb967", "generation_seconds_batch": 16.646875888924114, "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": 6.197352409362793, "genuineness": 4.419482707977295, "identity": 0.44531798362731934, "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_high/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ea1926ac916b8566e09fb3e1b74830d8b820c2d56b6d9f0a42381acedab53f9d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "b84e05792bf5b17363276b0d7744516f7640d5cafa33e645d5802f1450238005", "generation_seconds_batch": 24.20763007795904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6621131896972656, "target_z": 2.4062498567676704, "ce": 6.054384231567383, "genuineness": 4.036643981933594, "identity": 0.5196675062179565, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ea1926ac916b8566e09fb3e1b74830d8b820c2d56b6d9f0a42381acedab53f9d.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "ea1df4187a01eb4aae7bdaa20189e156e354502dd250af9351e6e6622dc88ecc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c3bcf07a1965cf2715a6802c041af8e59ae9da150fce9b263feb727802e3c088", "generation_seconds_batch": 23.326321093947627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.067453384399414, "target_z": 0.490234419760103, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.645686686038971, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Dis-moi quand tu es prêt.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ea1df4187a01eb4aae7bdaa20189e156e354502dd250af9351e6e6622dc88ecc.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "ea4f4dfe5628bd5dfaa569ccf6a030f1b0dc092a767b4d25af2a4cc452bce753", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "27ac9249c40477d965fbce7b0450c489110745ba444cffd8588ee6cfb5dacd56", "generation_seconds_batch": 17.299062434001826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8528814315795898, "target_z": 0.23242195556818546, "ce": 6.395919322967529, "genuineness": 4.030367851257324, "identity": 0.6518645882606506, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ea4f4dfe5628bd5dfaa569ccf6a030f1b0dc092a767b4d25af2a4cc452bce753.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "eab68c7585bed935b996c4a464f64c25bcf09fc2b776e14fdf2007b7bea096a4", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e8f33d59f520b2e7219b78064d2dfd51ff7b2aa3b2ac7b67ad27a7eaae1e6b0e", "generation_seconds_batch": 17.177780136931688, "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": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.6509673595428467, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/eab68c7585bed935b996c4a464f64c25bcf09fc2b776e14fdf2007b7bea096a4.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "eaeae066f3f13241cfb80fc775a43ec2b074cabb3d7a4de56cf6ba6fb560d120", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1322850bd46fdcbdc7d5570e0b4f6b12675ca896a9931ff5dab4c144e78c410a", "generation_seconds_batch": 21.68409943394363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.88214111328125, "target_z": 0.26757804443181454, "ce": 5.919358730316162, "genuineness": 4.6705241203308105, "identity": 0.6186126470565796, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/eaeae066f3f13241cfb80fc775a43ec2b074cabb3d7a4de56cf6ba6fb560d120.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "eb14eecf6e25f3e87e0506dbcd02b3f7875f277f326290a1848400a91dcbe8b0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d9cc310b6b073997a3f89ca1f6427fd41fe94427888e986e9d922d32848e3eb4", "generation_seconds_batch": 22.108564383001067, "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": 6.3641486167907715, "genuineness": 4.293961524963379, "identity": 0.41970330476760864, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/eb14eecf6e25f3e87e0506dbcd02b3f7875f277f326290a1848400a91dcbe8b0.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "eb4e7b325092bbaf494490e547e5b74a78d49ecf7e394c79f7224e81e081086d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "41837e4c28ddbd21c6678404509878f703607014e3ace7dc016d6b0cfd6b0ba4", "generation_seconds_batch": 18.523241147049703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.416945457458496, "target_z": 0.9101560888636291, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.7125424146652222, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/eb4e7b325092bbaf494490e547e5b74a78d49ecf7e394c79f7224e81e081086d.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": "eb6618f67944577bb9e2f671979130ffa0d0cca8613b70dc16d18be9656f7709", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "95aed391c694830db468463cb9911c5815323b6840ad3064cddcce5a0e46ab04", "generation_seconds_batch": 22.828968478017487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.790821075439453, "target_z": 1.3593750716161648, "ce": 5.911416053771973, "genuineness": 4.118232727050781, "identity": 0.5709329843521118, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/eb6618f67944577bb9e2f671979130ffa0d0cca8613b70dc16d18be9656f7709.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "9ac7824d72f25b2bccf6fcd6392dfad51239b7da224167c8f31e2016c5642d43", "generation_seconds_batch": 17.84025018394459, "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": 6.0146708488464355, "genuineness": 4.381826400756836, "identity": 0.8352036476135254, "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_high/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "ecf13fadea80bc4720c759312ebff4103570e7bab0c94bf52ad84c99c28a31d8", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5a9fa27439c0524e9d55a435c2e71487f937767c6cf5eac3d44a341a6399b750", "generation_seconds_batch": 22.13491012295708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.316161632537842, "target_z": 0.7890623209595878, "ce": 6.062326908111572, "genuineness": 4.7332844734191895, "identity": 0.7792500257492065, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ecf13fadea80bc4720c759312ebff4103570e7bab0c94bf52ad84c99c28a31d8.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "ed483f5a8be54d8c9465afda29aa30a94378aa139e2f0d7a67b96460a95ae3ad", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ca4c88e77c60d7adcc54f6ab92aedbb83ebb099e10d9b8f1e78a810b51a14f6b", "generation_seconds_batch": 23.962517819018103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.758310079574585, "target_z": 1.3203124641919175, "ce": 6.086154937744141, "genuineness": 3.6789097785949707, "identity": 0.5471746921539307, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ed483f5a8be54d8c9465afda29aa30a94378aa139e2f0d7a67b96460a95ae3ad.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "ed82fbf1b5527b946c053d0380f04a1a30cb063b8229bc2373755b01e207c47a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "1adff72e9594eb7125f4ec218111f051cfae496b175cce5d50016efda06c144c", "generation_seconds_batch": 16.58229361602571, "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": 6.213237762451172, "genuineness": 4.65797233581543, "identity": 0.6645422577857971, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand on est prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ed82fbf1b5527b946c053d0380f04a1a30cb063b8229bc2373755b01e207c47a.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "edb2f19fc6cd2f6601f84e696a3f793c6da41a75ab689766d2146a6f74d47ee6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "20def00f4014e9d62ff10f7ac380ae595f320fd638f30414eb8f628587a91dcb", "generation_seconds_batch": 22.538426212966442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.232969284057617, "target_z": 1.8906249283838352, "ce": 5.982900142669678, "genuineness": 4.124508857727051, "identity": 0.5513462424278259, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/edb2f19fc6cd2f6601f84e696a3f793c6da41a75ab689766d2146a6f74d47ee6.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "edce15b1dd7cbf35aca9d9f391608b97f542695b4cfcc1d46040fea50f09394a", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6f31a3e89576d526005ad383367072acc5cd1e0ec68c32b59827a955582590c3", "generation_seconds_batch": 22.809880690998398, "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": 6.3641486167907715, "genuineness": 4.582659721374512, "identity": 0.6164071559906006, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/edce15b1dd7cbf35aca9d9f391608b97f542695b4cfcc1d46040fea50f09394a.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": "ee4bb6799f6153b7fd1267eedee18e717240083eed9b9ae51718b6e329cfc9ba", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7f9a43be3f2088e25eeac9bcb1e14f49aad2c2408f31cf9c15ba5e7036d38c89", "generation_seconds_batch": 18.068316887016408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50606107711792, "target_z": 2.2187501432323296, "ce": 5.990842819213867, "genuineness": 4.093128204345703, "identity": 0.8368833065032959, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Et je bécle après, je t'attendrai de l'autre côté. J'appelle.", "wer": 0.25, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ee4bb6799f6153b7fd1267eedee18e717240083eed9b9ae51718b6e329cfc9ba.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ef3a6be22d14bf57224464f474df7e4e31c6959ea8bc039df74bc1c39fb3f46e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "331d6b6aee1365de612d78d1997ecb312c88aea0a33bfa9b1636019c638e118d", "generation_seconds_batch": 23.33582736097742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.01323127746582, "target_z": 2.8281252148484946, "ce": 6.0146708488464355, "genuineness": 4.13078498840332, "identity": 0.4287312626838684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ef3a6be22d14bf57224464f474df7e4e31c6959ea8bc039df74bc1c39fb3f46e.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ef9df171b1d183336d5ea73df9684fcfd7d1476038156811355b73ba31957a89", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "f366ec957cd3dc79c208374cefb9a49926de0b57862c0d67897d6a898a695bc7", "generation_seconds_batch": 17.053619588026777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.284986734390259, "target_z": 1.9531249283838352, "ce": 6.0305562019348145, "genuineness": 4.093128204345703, "identity": 0.4685804545879364, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ef9df171b1d183336d5ea73df9684fcfd7d1476038156811355b73ba31957a89.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "efa3a1964e43604227ee91efee00bc837a1db6819beeaf5644f1b7b7a43e3bf6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "835e25f7fd6c64cd3e56a6799e5e9a21dc0863291eeb4c6b01f1b73270aa2457", "generation_seconds_batch": 23.4522893599933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4800522327423096, "target_z": 2.1875, "ce": 5.534139156341553, "genuineness": 4.024092197418213, "identity": 0.5382726192474365, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime, ¿cuándo estás listo?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/efa3a1964e43604227ee91efee00bc837a1db6819beeaf5644f1b7b7a43e3bf6.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "efcea5ab9965f4430c996334a4e36e09c9ec022c7481c3cce4660debfcec6974", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "50d5b41bf820f9c93abf741bda98d182da712f15fad0001332bcfcf300babcb8", "generation_seconds_batch": 22.94138723402284, "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": 6.332377910614014, "genuineness": 4.469690799713135, "identity": 0.514204740524292, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/efcea5ab9965f4430c996334a4e36e09c9ec022c7481c3cce4660debfcec6974.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "f00d0abee3ef2988d646f93b8493d619937a7cd4423481e2b99ce5877432f7bc", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "502a7025f21dfa8ec0f98541456a37ac354e78d2c54a137e1eebd350b2ec7ff9", "generation_seconds_batch": 16.973177323001437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680284023284912, "target_z": 1.2265626074242473, "ce": 6.403861999511719, "genuineness": 4.331617832183838, "identity": 0.3606959581375122, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f00d0abee3ef2988d646f93b8493d619937a7cd4423481e2b99ce5877432f7bc.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "f03fa36166cabb222706941d0e04dd82dbb35d629823fd8460feb56cf9a9f33e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "eefb25d02057fadeb7d9e6ca2c8d1d6a3a1d4ccc1ea4ba018bb40c52be20ca43", "generation_seconds_batch": 18.024170089047402, "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": 6.2767791748046875, "genuineness": 4.507347106933594, "identity": 0.6830185651779175, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f03fa36166cabb222706941d0e04dd82dbb35d629823fd8460feb56cf9a9f33e.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": "f08603c1004c173ed5b4792e6043f45152715000898d517592f23bc14829b1ba", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "095876ffb74933cbf749e348241b81cc81883988d4f43ce0f9d038c55e209b10", "generation_seconds_batch": 17.47242779203225, "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": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.8255825042724609, "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_high/f08603c1004c173ed5b4792e6043f45152715000898d517592f23bc14829b1ba.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "f09aa360dae77ffc2e93e297ee9fb40bf67640b907d35e15bf8a16a23be945f6", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9de9865a21be8517db2b781bb11946a3058ef3f95c0e7a1ba0ace23673211d7a", "generation_seconds_batch": 22.005086333025247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6621131896972656, "target_z": 2.4062498567676704, "ce": 5.855817794799805, "genuineness": 4.174716949462891, "identity": 0.6857221126556396, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f09aa360dae77ffc2e93e297ee9fb40bf67640b907d35e15bf8a16a23be945f6.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "f1352b1d6b3bd51438b121f535c90e81895f447f3c92d8f28175443615a2825f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "fa25fb96f5dc34524a5ec35cb6d8440fd6e968e622e5b9f6e455ae77eb315c32", "generation_seconds_batch": 18.668961566057988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2283823490142822, "target_z": 0.6835937679040412, "ce": 6.268836498260498, "genuineness": 4.7332844734191895, "identity": 0.8161482214927673, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f1352b1d6b3bd51438b121f535c90e81895f447f3c92d8f28175443615a2825f.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "b28c49370f77f635394dfe976a0d772987b04a3a84728c66708f966cf9b69a63", "generation_seconds_batch": 17.169369822950102, "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": 6.23706579208374, "genuineness": 4.394378662109375, "identity": 0.7613987922668457, "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_high/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f226e5d5212ae064d2f9a0aa70e00fdbeed3001dd8e0f5e1fcc0194af0a4ebb4", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8be470c01d2876671ea5257d13677282680d2fb4e3a90c3b102fd1ece6ff2f3e", "generation_seconds_batch": 17.59279162599705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6177868843078613, "target_z": -0.05004881357796651, "ce": 6.435632705688477, "genuineness": 4.168440818786621, "identity": 0.6372041702270508, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f226e5d5212ae064d2f9a0aa70e00fdbeed3001dd8e0f5e1fcc0194af0a4ebb4.mp3" }, { "id": "vn_TENS_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_high-en-1", "key": "f22ff01ded25ab143c5b6e8f661acda918a7fd72cf0a223a32d1235395fa07cd", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c6a5784599bc1568590ddbee5939bcf0777dcd5670b277979839672cd93b108d", "generation_seconds_batch": 17.86357043602038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7781063318252563, "target_z": 0.14257818766414426, "ce": 6.443574905395508, "genuineness": 4.582659721374512, "identity": 0.30668479204177856, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f22ff01ded25ab143c5b6e8f661acda918a7fd72cf0a223a32d1235395fa07cd.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "f249819d318824960bfcb51a3aa7485b3ead07082335bcf942bbfd87a204ba69", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "6490006be0d352f9bb649e3e95cb390716a41bbca3b757011f6a4b2ad5dca8f6", "generation_seconds_batch": 23.326321093947627, "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": 4.369274139404297, "identity": 0.6871949434280396, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f249819d318824960bfcb51a3aa7485b3ead07082335bcf942bbfd87a204ba69.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "f291aa5173c6e44ec783fef778758863ab2816ec7bfe5b443dc0c5bceb7430b2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8727e63f6a50a1a3f2ca86ef876b0552bb9536eb78e36fd42b1abba93b4d88ad", "generation_seconds_batch": 23.396226670010947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3714303970336914, "target_z": 0.855468839520206, "ce": 6.197352409362793, "genuineness": 4.6956281661987305, "identity": 0.6854923963546753, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f291aa5173c6e44ec783fef778758863ab2816ec7bfe5b443dc0c5bceb7430b2.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "f2996a14ead53885765c4657670371583d6ce1a84317006eb2deb1db97332385", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "80bf54d0f815c4e01940c6f4dbe3027d4a63aa666af6d15b1a0b4bf0b8a26a17", "generation_seconds_batch": 16.489943793974817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.117845296859741, "target_z": 0.5507813037121236, "ce": 6.284721851348877, "genuineness": 3.9676074981689453, "identity": 0.3268831670284271, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f2996a14ead53885765c4657670371583d6ce1a84317006eb2deb1db97332385.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "f2ed00c9a789b25ad7475e4b4021dbe763a79cb7c439f71f21868cb5c9d28493", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "609d53e33fe4ef928e9a8a9bf6a7dc57c73d10519eb66bea78e6a8aa04965ab2", "generation_seconds_batch": 24.200259198085405, "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": 6.149696350097656, "genuineness": 4.821148872375488, "identity": 0.7133347392082214, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f2ed00c9a789b25ad7475e4b4021dbe763a79cb7c439f71f21868cb5c9d28493.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "f2f9452a46d9d09016d751de27facbe90dadbdc26fb3d1ea3b516be53fe60aef", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "399a08510d197082a5c0467221701bedf0225375077d87310293663075aa335f", "generation_seconds_batch": 21.851496356073767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9341585636138916, "target_z": 0.33007804443181454, "ce": 6.213237762451172, "genuineness": 4.758388519287109, "identity": 0.8497954607009888, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f2f9452a46d9d09016d751de27facbe90dadbdc26fb3d1ea3b516be53fe60aef.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f332f4eb2ba22170241ac84580042c51bbfdb9e8482d9b09b85f5c0417f12b1a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "86fa2acd774b4e612070ca8df35ebdc8dc8aca31f8ce4f8a63654ff504617772", "generation_seconds_batch": 16.184966755099595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6402397155761719, "target_z": -0.023071290741003864, "ce": 6.459460258483887, "genuineness": 4.306513786315918, "identity": 0.5633032321929932, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f332f4eb2ba22170241ac84580042c51bbfdb9e8482d9b09b85f5c0417f12b1a.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": "f38bd2c38b8c2ea8131f9a6627d6b5b0581f0c6ff41e5b756dff2f7e2d2c3656", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "940ebeb2e5bfc309b61bd4e873ccd34253fdc323a90b9c281edcd6de8f8cc85a", "generation_seconds_batch": 23.042630922980607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680284023284912, "target_z": 1.2265626074242473, "ce": 6.268836498260498, "genuineness": 4.256305694580078, "identity": 0.6535919904708862, "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_high/f38bd2c38b8c2ea8131f9a6627d6b5b0581f0c6ff41e5b756dff2f7e2d2c3656.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "f3d289b99c72cb5c1c0285b2d517f990a8dd15412cfcb6e651f3f28f03774c9f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "da3dbd192debd534f33f2e28215758c0e8a8c79dbc746cc88649e025e0f9b179", "generation_seconds_batch": 17.28753367101308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2654802799224854, "target_z": 1.9296875358080825, "ce": 6.316492557525635, "genuineness": 4.469690799713135, "identity": 0.5984105467796326, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f3d289b99c72cb5c1c0285b2d517f990a8dd15412cfcb6e651f3f28f03774c9f.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": "f43bf12f95a74637735d478aa35b956ef6a64e1518c9a1205ea91bcac034d2d7", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6f83fa849dd7ccca7f348e67459c0c382b98a35afd5e9eee513ad63c48a2fb5", "generation_seconds_batch": 17.026644231984392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4800522327423096, "target_z": 2.1875, "ce": 5.824047088623047, "genuineness": 4.024092197418213, "identity": 0.556551456451416, "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_high/f43bf12f95a74637735d478aa35b956ef6a64e1518c9a1205ea91bcac034d2d7.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "f44ed100ad45c4c9316b545658da2a9d503ac8143e04cf0e019bd2b8ecd2f405", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "28f7a0fd15d97408807acacacd911312330b661fd1b7921827535e38b267a07d", "generation_seconds_batch": 21.076287613017485, "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": 5.990842819213867, "genuineness": 4.457139015197754, "identity": 0.5520768165588379, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f44ed100ad45c4c9316b545658da2a9d503ac8143e04cf0e019bd2b8ecd2f405.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "f4c0fe93d6c81416611cb4516e9f61e9ef4cb2d79f37632d0fecdbd56ef4a049", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ffb2f1a52b77d63b5ac2d6404bac5b9d65a81ce91ac100936ae2075f77cbbead", "generation_seconds_batch": 22.050309470971115, "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": 6.0146708488464355, "genuineness": 4.507347106933594, "identity": 0.4953041672706604, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f4c0fe93d6c81416611cb4516e9f61e9ef4cb2d79f37632d0fecdbd56ef4a049.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f4c5e08422987480b237fcdb75539b1a7b86207ec8daac7bc8982175e8c56875", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "8b871a9c5b061e5a25832151a621878453176c112a0c715d15543fb94f392970", "generation_seconds_batch": 22.48584399803076, "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": 4.632867813110352, "identity": 0.3257758319377899, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f4c5e08422987480b237fcdb75539b1a7b86207ec8daac7bc8982175e8c56875.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": "f4cf86e1b4e4fae5bbc5d45437b6e3869b030621ab35f6b449cee66f4ded1cc2", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6b204c4eb83f5a11c86975743fcb7aa6981719a3dba31a8e5b1921c4f5d91bc", "generation_seconds_batch": 23.135435128002428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6672797203063965, "target_z": 1.210937679040412, "ce": 5.903473377227783, "genuineness": 4.218649387359619, "identity": 0.6911357045173645, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lisez le bien, et vérifie les deux derniers chiffres. Appelle moi quand tu auras un moment.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f4cf86e1b4e4fae5bbc5d45437b6e3869b030621ab35f6b449cee66f4ded1cc2.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "f547b658e43745478e528bba34eed0b1ab2c2d827ff7d780dd885b87b76e38a4", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "e849fd17f7ea5eec88d6dfedbecf87f487841c8933f4162e8eab56d85c140aed", "generation_seconds_batch": 22.75012099801097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5177292823791504, "target_z": 1.0312498567676702, "ce": 5.911416053771973, "genuineness": 4.469690799713135, "identity": 0.3525660037994385, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f547b658e43745478e528bba34eed0b1ab2c2d827ff7d780dd885b87b76e38a4.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "f65bfd0262daf80bb18ff039bf3f53fb785d03ea7075944575bd1272e64dafa7", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "9f922548d251dcd054d4b06aa2541296763fa0c19c31daf9dc86bc67286941d3", "generation_seconds_batch": 18.754759349976666, "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": 6.054384231567383, "genuineness": 4.796044826507568, "identity": 0.8180433511734009, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f65bfd0262daf80bb18ff039bf3f53fb785d03ea7075944575bd1272e64dafa7.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "f66a4da69a5f8d705d00b1be422a5a265c1ca6c239f76fe411cfa087fd7c3376", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9f29f01387c299b164daa2755dad4d2433da812482af74b3a2482268d51dfa57", "generation_seconds_batch": 23.781660664943047, "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": 5.919358730316162, "genuineness": 4.394378662109375, "identity": 0.4600571393966675, "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_high/f66a4da69a5f8d705d00b1be422a5a265c1ca6c239f76fe411cfa087fd7c3376.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "f692a91d4d7e738181aada2053af1a4d08fcb2441e53dc44725c463b614813ae", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "5d8d3447d4f963473c785bf1903906f91b59e485cdc353d29e37a40170d770d7", "generation_seconds_batch": 22.28453195595648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117265701293945, "target_z": 2.953124641919176, "ce": 5.5539960861206055, "genuineness": 4.055472373962402, "identity": 0.5100696086883545, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f692a91d4d7e738181aada2053af1a4d08fcb2441e53dc44725c463b614813ae.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "f6df7597305e51dac4a17d09dd1303fccbc498275f7aa34bf39c3be40d2e72b9", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "87fc4391ed3bec811eaa31f57936921fb1cd7613e7aafa12511b6aff1c886ee5", "generation_seconds_batch": 17.714737994014286, "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": 6.24500846862793, "genuineness": 4.532451152801514, "identity": 0.5156750679016113, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f6df7597305e51dac4a17d09dd1303fccbc498275f7aa34bf39c3be40d2e72b9.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "f75f10729bfe536a7f5c42c3208ae7a2b7fb5b325f1304211106e12d6ed240f1", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3e24e9a2f07e3d3ab1681a2e6157d921b10916fa58e1a6553084b7a847922d29", "generation_seconds_batch": 16.489052001968957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8138682842254639, "target_z": 0.1855468839520206, "ce": 6.054384231567383, "genuineness": 4.356721878051758, "identity": 0.6004654169082642, "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_high/f75f10729bfe536a7f5c42c3208ae7a2b7fb5b325f1304211106e12d6ed240f1.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "14e3f18ebda60077593bb96f9157b7043136f7f1b4b8c6f60933b32cd00a7940", "generation_seconds_batch": 17.888495799037628, "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": 6.213237762451172, "genuineness": 4.808597087860107, "identity": 0.6659567356109619, "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_high/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.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": "f7a97cf5a950a5387f57eb1c1eb9d9ee4e5e7d904a3c12fa6882a6c41121eb2b", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f689ffdf7b38639a3a5d85c77ce625c6b459604d9347cde09b51d0762fa1a8a6", "generation_seconds_batch": 23.030612318078056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0834193229675293, "target_z": 1.710937679040412, "ce": 5.732706069946289, "genuineness": 4.262581825256348, "identity": 0.6314589977264404, "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_high/f7a97cf5a950a5387f57eb1c1eb9d9ee4e5e7d904a3c12fa6882a6c41121eb2b.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "f84581a6ba2ff607da269bd452e7b128bd16ff86eb2e28202a56206ef3ddaf3f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "4eb6d28726a5044b18cb82afcd73341d234c37f79fbe23b0a7443a89d848fd8b", "generation_seconds_batch": 24.813554077991284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049572467803955, "target_z": 0.4687501432323297, "ce": 5.855817794799805, "genuineness": 4.306513786315918, "identity": 0.5521735548973083, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f84581a6ba2ff607da269bd452e7b128bd16ff86eb2e28202a56206ef3ddaf3f.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f85ce18eb7234e63fecccd48ade0429cfe1ba397b252a48ac254ee64f279c39d", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "db61cce4237f5977826607608b2f80dda92d3970ac916c1171015f3e3102b196", "generation_seconds_batch": 23.61446019797586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0073082447052, "target_z": 0.4179688395202061, "ce": 6.284721851348877, "genuineness": 4.3881025314331055, "identity": 0.44641411304473877, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f85ce18eb7234e63fecccd48ade0429cfe1ba397b252a48ac254ee64f279c39d.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_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": "a8889ae916a69d07e32cbb327af47390a948917d2f645df55a46de3883b9f4a2", "generation_seconds_batch": 16.118621718953364, "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": 6.300607204437256, "genuineness": 4.406930446624756, "identity": 0.5686091184616089, "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_high/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.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": "f88f9170bca139a8afd7577fa77df8375b5ae331717a2add286bee354b9118ea", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ba561553c41e6d962af9fce3bf3731727226111190d3c11a8426c83cd8244d7", "generation_seconds_batch": 18.423899870947935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8818516731262207, "target_z": 1.4687501432323298, "ce": 6.173524379730225, "genuineness": 4.532451152801514, "identity": 0.5448846817016602, "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_high/f88f9170bca139a8afd7577fa77df8375b5ae331717a2add286bee354b9118ea.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f8bc0d26af77ec76e17d8b617b817533205bfd9ba89561d66ebe4e31ef370b7b", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "a82bb947806135821fdd470226572894492d74fe46c2dc43e57e90dc819174b9", "generation_seconds_batch": 25.796130744973198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7634763717651367, "target_z": 0.125, "ce": 6.427689552307129, "genuineness": 4.7834930419921875, "identity": 0.6397796869277954, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f8bc0d26af77ec76e17d8b617b817533205bfd9ba89561d66ebe4e31ef370b7b.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "f92323cc9b14d1a29929f0771362aeaa4980cb4e0951786394b567e785302fe7", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c3dea04bd1897e5fb7d2f59345f8991166ab656cc79c21f6ff5a155b0c8b1a22", "generation_seconds_batch": 25.0606854299549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.864260196685791, "target_z": 0.24609376790404122, "ce": 5.911416053771973, "genuineness": 4.557555198669434, "identity": 0.7688792943954468, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f92323cc9b14d1a29929f0771362aeaa4980cb4e0951786394b567e785302fe7.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "f92526aeccbb7c8a2cd738a9d85bba90d66acff72f0e9522530288b832af9899", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "0dc288085e51abe9f677df712d7ab28844f3cf47fd49a5ba5ab3691250746027", "generation_seconds_batch": 17.48535330593586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.764289140701294, "target_z": 0.1259765580239897, "ce": 5.863759994506836, "genuineness": 4.293961524963379, "identity": 0.5346782207489014, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f92526aeccbb7c8a2cd738a9d85bba90d66acff72f0e9522530288b832af9899.mp3" }, { "id": "vn_TENS_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_high-de-1", "key": "f97b70be6bc3f7f91b8fdc8000fcfe2f8b4d80196cb5a3bf6daa12e49e760b88", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d371879a8ffc88db47f82ab7f0c92ed8e6e0925fb88cbfd1d95c433fffa985ac", "generation_seconds_batch": 16.630346002988517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8488174676895142, "target_z": 0.22753902221590727, "ce": 6.475345611572266, "genuineness": 4.469690799713135, "identity": 0.5130212306976318, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/f97b70be6bc3f7f91b8fdc8000fcfe2f8b4d80196cb5a3bf6daa12e49e760b88.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "f98380664b45aa8af4f464162ee4d386ae8d7bd1e0b02ee11a39185830608e1d", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f4e3de1566845bebe01116d575f46d1237c36f276340e798e12c84f632e3e435", "generation_seconds_batch": 17.08283402491361, "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": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.7231889367103577, "hypothesis": "De deur 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_high/f98380664b45aa8af4f464162ee4d386ae8d7bd1e0b02ee11a39185830608e1d.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "fa03d69939e8e5881dc26deec1c9170205648aa5ea903f3d76e8cd1b2e35413a", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "430cab9204b1638e6be87a56da08aef2706d79d328e48cc3b58eb4f9d429d163", "generation_seconds_batch": 16.845726281986572, "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": 5.895530700683594, "genuineness": 4.231201171875, "identity": 0.6549637913703918, "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_high/fa03d69939e8e5881dc26deec1c9170205648aa5ea903f3d76e8cd1b2e35413a.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "fa8b96e78d8a8c41a7f8282d1f5bd6a196fbb0687ec2ce95d1a16679df9546df", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "ee64b3dbe5ff97f8a8f3b385cc2a6fa295927c95cde8ca0f97956a67ba6789ac", "generation_seconds_batch": 21.794991574948654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.481388092041016, "target_z": 3.390624928383835, "ce": 6.197352409362793, "genuineness": 3.8169827461242676, "identity": 0.5800105333328247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fa8b96e78d8a8c41a7f8282d1f5bd6a196fbb0687ec2ce95d1a16679df9546df.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "fb141686b88a79830711f40efc2352cc2dc2d789d82d04354c99f04549d9e48f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "7d531aa4c2c8f823dc7d9ed70e596cb42b24ddff3eac57cce7220002c823e7fc", "generation_seconds_batch": 23.1312564809341, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.702462196350098, "target_z": 3.6562498567676704, "ce": 5.692992687225342, "genuineness": 3.625563621520996, "identity": 0.3867748975753784, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fb141686b88a79830711f40efc2352cc2dc2d789d82d04354c99f04549d9e48f.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "fb4f17d5432436f6f4399bd920ef37b8fe84fe8225678f6fe679e205b121bda0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "6b47ad05f98373305a6174488095e3410564618e9fd2daf592762e11a95a8155", "generation_seconds_batch": 22.525076239951886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8818516731262207, "target_z": 1.4687501432323298, "ce": 5.617537498474121, "genuineness": 4.256305694580078, "identity": 0.2782108783721924, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fb4f17d5432436f6f4399bd920ef37b8fe84fe8225678f6fe679e205b121bda0.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "fb597228a322779f8d6c9ce63fef1bc804f7d689032acf40573497874d495808", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "43ebc5b6f8aca50ac4f7bcfd8a101218cf5ea0535eddbc8b28dd773bf11c2ed1", "generation_seconds_batch": 23.63948260503821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7401394844055176, "target_z": 2.5, "ce": 6.173524379730225, "genuineness": 4.143336772918701, "identity": 0.5283650159835815, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fb597228a322779f8d6c9ce63fef1bc804f7d689032acf40573497874d495808.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "fbab3b88ae68d5172e18ca69fc7eb174188d486fd9581b7ace4c0aeb480142d0", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "281d3cdc3ae27fa6c2b0d4623bd46be6e29b8cd164875fadf82ba3a2131fffab", "generation_seconds_batch": 23.302728619077243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8065533638000488, "target_z": 0.17675786173611333, "ce": 6.213237762451172, "genuineness": 4.042920112609863, "identity": 0.3880927562713623, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fbab3b88ae68d5172e18ca69fc7eb174188d486fd9581b7ace4c0aeb480142d0.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "fc77d4f6ab3774f6a55c2d912bf32631070220ac71cb17d50e0d833517dcd6e2", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "c0b10ce062eb34a084d0f85bb167ec891b8accf74db0b7296d44f76607001565", "generation_seconds_batch": 23.92921092105098, "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": 5.585766792297363, "genuineness": 4.055472373962402, "identity": 0.5443341732025146, "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_high/fc77d4f6ab3774f6a55c2d912bf32631070220ac71cb17d50e0d833517dcd6e2.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "fca281b55cb59e57dac2a470d9ee9e5b11345cced4afa0115da83ccafef26d7f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "768de68f263b82949a39430e1c711c4f2f8636cf5f9f67c184893dd5eb69c285", "generation_seconds_batch": 16.75442807096988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0056827068328857, "target_z": 0.41601572347222665, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.4691812992095947, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fca281b55cb59e57dac2a470d9ee9e5b11345cced4afa0115da83ccafef26d7f.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": "fcb609d43f694fee77ec9fb2316df33b2f19e96c81ff5cb0db1a413314c0ff07", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18470749e032f3a41bc4123a3cf8fbf740a9f54a0df610a0c9ca3315e0920cdd", "generation_seconds_batch": 17.14151410292834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868847131729126, "target_z": 1.4531249283838352, "ce": 6.24500846862793, "genuineness": 4.281409740447998, "identity": 0.4706820845603943, "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_high/fcb609d43f694fee77ec9fb2316df33b2f19e96c81ff5cb0db1a413314c0ff07.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "fd1761ef82e2b50d961e85fc79b1d4ea9612b56c39e59d324b6e43b95eff2b38", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "8adfa756bd11fee239f7d71536f0e83f6e5b4829c6b5ca4211165bbd873d7e36", "generation_seconds_batch": 22.871165193035267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1549432277679443, "target_z": 1.7968750716161648, "ce": 5.019851207733154, "genuineness": 4.162164688110352, "identity": 0.7256436944007874, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fd1761ef82e2b50d961e85fc79b1d4ea9612b56c39e59d324b6e43b95eff2b38.mp3" }, { "id": "vn_TENS_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_high-de-0", "key": "fd8ac959edb66a2f399490b7ab90c9891d6e56f640606005ca2a3f492f7eb89e", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "d6de3e65ade24845e595643fa6281a46cc2846982007472889a6b5103f715c2e", "generation_seconds_batch": 22.84008495497983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6672797203063965, "target_z": 1.210937679040412, "ce": 5.871703147888184, "genuineness": 4.457139015197754, "identity": 0.5847649574279785, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fd8ac959edb66a2f399490b7ab90c9891d6e56f640606005ca2a3f492f7eb89e.mp3" }, { "id": "vn_TENS_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_high-fr-1", "key": "fd9dbfd3f540219e460fe73e566d1c8e87e63dc6acad4851e45ce06c27a5d659", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ae14b5d7cda4a92fcf4a473dff2a2601fe75d075b5b482929a73fed8a71852b5", "generation_seconds_batch": 24.126250863075256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8528814315795898, "target_z": 0.23242195556818546, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.7046008110046387, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fd9dbfd3f540219e460fe73e566d1c8e87e63dc6acad4851e45ce06c27a5d659.mp3" }, { "id": "vn_TENS_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_high-fr-0", "key": "fdab63bbae7afba96bb5c0d7ed708d14fd5081017f08174318fa7535e19fe409", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "157ba68a4df17312a2fc175092891017c463c77070530969d4f49483ef2a65d7", "generation_seconds_batch": 18.287825005943887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1549432277679443, "target_z": 1.7968750716161648, "ce": 6.260893821716309, "genuineness": 4.582659721374512, "identity": 0.7421314120292664, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Dis-moi quand tu es prêt.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fdab63bbae7afba96bb5c0d7ed708d14fd5081017f08174318fa7535e19fe409.mp3" }, { "id": "vn_TENS_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_high-en-0", "key": "fe49951a57e836f0f30240510887441e6d95fb8f0ee7df3372f409c96dd5e728", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "dc596b628f2ce71ddc195a3155e9090ea6e78e0075cfe7c8dd0a7eba3236babe", "generation_seconds_batch": 23.26225476397667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.050908327102661, "target_z": 1.6718750716161648, "ce": 5.998785495758057, "genuineness": 3.973883628845215, "identity": 0.6357278823852539, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fe49951a57e836f0f30240510887441e6d95fb8f0ee7df3372f409c96dd5e728.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": "fe4daece85917af3dff07773fb55304f2bdbcac83e5db7289c3ec660b747df8c", "bucket": "vn_TENS_reg_high", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba1adaf79d47e4d00202c7515b652e9fba55b8d864fa5c8265ee622b41505ec0", "generation_seconds_batch": 22.667473532957956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7453057765960693, "target_z": 1.3046875358080825, "ce": 6.133810997009277, "genuineness": 4.632867813110352, "identity": 0.662940502166748, "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_high/fe4daece85917af3dff07773fb55304f2bdbcac83e5db7289c3ec660b747df8c.mp3" }, { "id": "vn_TENS_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_high-es-1", "key": "fe844c448f43de82e2e0ec5e1c45d8e42799a66d066b10a89f5253afcf8eff64", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "90ce2e99077303a9dae923f5fa92d163a8c22bfd6c8dc0d38a2d826a7ee4658b", "generation_seconds_batch": 22.661905245040543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5269595384597778, "target_z": -0.1591796740719691, "ce": 6.316492557525635, "genuineness": 4.65797233581543, "identity": 0.5450766086578369, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/fe844c448f43de82e2e0ec5e1c45d8e42799a66d066b10a89f5253afcf8eff64.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "ff7f0db8d9fe447ad57307a971a0236e29b2a23ac497a6ba9e490a7f76eebcab", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": 127, "seconds_generated": 10.16, "mp3_sha256": "11461b855611f412fd02fe1c9e6ecdf563356fb1ddf4f67e864c0bbececf9fb0", "generation_seconds_batch": 23.12853293691296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.909196138381958, "target_z": 2.703124928383835, "ce": 5.395142555236816, "genuineness": 3.8797430992126465, "identity": 0.4980587959289551, "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_high/ff7f0db8d9fe447ad57307a971a0236e29b2a23ac497a6ba9e490a7f76eebcab.mp3" }, { "id": "vn_TENS_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_high-nl-1", "key": "ff92f9be3fa8ff5933f865595582b7074fd94fef6c4e92a4a6933e0e834ccb9c", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "2fdb92a75a345dbef31c7d3f633eabdb6e9b8feacb17f58809c1ff5ddf686400", "generation_seconds_batch": 17.959557310910895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7453057765960693, "target_z": 1.3046875358080825, "ce": 6.046441555023193, "genuineness": 4.124508857727051, "identity": 0.7085868120193481, "hypothesis": "Ik heb een klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ff92f9be3fa8ff5933f865595582b7074fd94fef6c4e92a4a6933e0e834ccb9c.mp3" }, { "id": "vn_TENS_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_high-es-0", "key": "ffb802a8f3220b6a3f2617d7f01d0e1420c70c6f6d856754ac1f74531121f98f", "bucket": "vn_TENS_reg_high", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "90a93a7620de336f05c2eb91d525985ecd6af6e975541069bab3102150702af9", "generation_seconds_batch": 22.637672243989073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9666695594787598, "target_z": 0.3691406518560618, "ce": 5.7247633934021, "genuineness": 4.080576419830322, "identity": 0.45250147581100464, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ffb802a8f3220b6a3f2617d7f01d0e1420c70c6f6d856754ac1f74531121f98f.mp3" }, { "id": "vn_TENS_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_high-nl-0", "key": "ffd649f4087dd7dfe84ab2e70fadc4bdc2ec1dfd4cbbbb72bae8ae13a6825aed", "bucket": "vn_TENS_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled articulation. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Audibly taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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 taut and strained delivery, with pressed breath, held muscular tension and tightly controlled 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": "29cb128901e2ddb2c7599d35d1fbca49854d50da707b1b2262584de0d38c3864", "generation_seconds_batch": 17.00179603707511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5177292823791504, "target_z": 1.0312498567676702, "ce": 6.0146708488464355, "genuineness": 4.1998209953308105, "identity": 0.652306318283081, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_TENS_reg_high/ffd649f4087dd7dfe84ab2e70fadc4bdc2ec1dfd4cbbbb72bae8ae13a6825aed.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "5b63ab26b85f0413b4ed41c9860cf045dd11b7b87f251b22a44cc0c0a6e3b4e9", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.9208645820617676, "target_z": 1.5156249283838352, "ce": 6.109982967376709, "genuineness": 4.425758361816406, "identity": 0.7061455845832825, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/6288b1d6395e976190e66ae3.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "0c18ed3e0754e4be7e36a3674a657a25bca812fdbc4c970a8a37abb0f5bf9b68", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.245973587036133, "target_z": 1.9062498567676702, "ce": 6.1020402908325195, "genuineness": 4.105680465698242, "identity": 0.5974409580230713, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/3da58a0df1a88bc22b074238.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "1ec78fdb2d901112226a3a473a09108c3ba919f834000e92fc4cf7c391ff2c28", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.3941879272460938, "target_z": 0.8828124641919176, "ce": 6.284721851348877, "genuineness": 4.419482707977295, "identity": 0.6687343716621399, "hypothesis": "Die Tür 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/audio/vn_TENS_reg_high/de/1c5bb20aa44d64fc406bf6f7.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "5bbe13819ab9a64baa5e3e65ab43c29107cbbd5d9d540abb72bb10fb1bfe728e", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1159300804138184, "target_z": 1.75, "ce": 6.2767791748046875, "genuineness": 4.507347106933594, "identity": 0.7344229221343994, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/9b7a1345f201f6bc8df7ad06.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "aefff5900699e6b7767b065bbe40ce85a7fc3169059a6df71b2587453b1d7fe5", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.018397331237793, "target_z": 1.6328124641919175, "ce": 6.157639026641846, "genuineness": 4.607763767242432, "identity": 0.7079563140869141, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/03e26a04f2104a4e98315836.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "6f770dde5d64ba35bd55be265623302ce37994726080d5150e357486f49fcd6e", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.420196533203125, "target_z": 0.9140623209595878, "ce": 6.221180438995361, "genuineness": 4.3881025314331055, "identity": 0.31375136971473694, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/b684b878fcd21618748a8b14.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "7b18bf7b91a2ec2cc4aa1da00930c9614184609e943ca777954d9f1f2494a94d", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.537235975265503, "target_z": 1.0546875358080825, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.4334971010684967, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/a2f7c893c126712b96974f97.mp3" }, { "key": "original-vn_TENS_reg_high-de-0-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "6465628d8c8f81d1115655c2ea5e96fd711841c86c40ed109bc4849fa10d8a8d", "task_id": "vn_TENS_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7401394844055176, "target_z": 2.5, "ce": 6.062326908111572, "genuineness": 4.030367851257324, "identity": 0.42022478580474854, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/71f5c4e24f64ad5cdeaaf0d7.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "70c587efc41051c54e01be355d6612877a8ed904c3e2ee3a72e3bb0fdc040a01", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049572467803955, "target_z": 0.4687501432323297, "ce": 6.24500846862793, "genuineness": 3.9173994064331055, "identity": 0.6612963676452637, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/c1c85ae9f3d3ce253bce78c6.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "50078ea4326b9801fe945a1b23e34ed61bd85a535f8a79371b67953b352dbecb", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0788321495056152, "target_z": 0.5039062320959588, "ce": 6.443574905395508, "genuineness": 4.356721878051758, "identity": 0.6922775506973267, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/c88d7c0ba45f293cdc358957.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "1dfa5c5d320e9b92a5c454d4912800fd3d41c4c095a28cafb291a2d519faabb2", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205624580383301, "target_z": 0.6562498567676703, "ce": 6.316492557525635, "genuineness": 4.419482707977295, "identity": 0.48912450671195984, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/cdf6bb9eb374c1b2dcf91858.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "7965bbe17696bb71704c06220d226a7b1a9f8310c4c1f26af2565f5e3e5ce7d6", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.025189161300659, "target_z": 0.4394531160479794, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.5951191186904907, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/a60dfbbf15694e3d43fe1d49.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "be492c86525e4bf4aec6b5bdb03940cd4f143de22ab44cf41093f8b2607d4157", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.60225772857666, "target_z": 1.1328124641919175, "ce": 6.340320587158203, "genuineness": 4.080576419830322, "identity": 0.5006933212280273, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/81fd710a63f266a4c941f237.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "ce5afd29a7d578ded52ea85c29ecc38e33801943dd3e5fbb3ed7a511bc0034d9", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0073082447052, "target_z": 0.4179688395202061, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.4295377731323242, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/00037569c8a1ea17293be904.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "12a33d4d0711a3a58f4879bb1c9986e51aebd8123b5027136df3604d78d8bf3d", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.9325330257415771, "target_z": 0.32812492838383517, "ce": 6.38797664642334, "genuineness": 4.300237655639648, "identity": 0.47135019302368164, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/c33d4f96d8de891e4b470d2c.mp3" }, { "key": "original-vn_TENS_reg_high-de-1-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "73f660d54dd7cc24ebbbe247a286fec0fb5e559c2bf855d7ad187cdaeafb25cd", "task_id": "vn_TENS_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.45920991897583, "target_z": 0.9609376790404122, "ce": 6.284721851348877, "genuineness": 4.068024158477783, "identity": 0.32667577266693115, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/de/b83269bbad88583c5d0f7cda.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "1eb71087956baeb9b5a1535f8020efae18740622923b5f4af724bbcb5567f919", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2251312732696533, "target_z": 0.6796875358080824, "ce": 6.205295085906982, "genuineness": 4.017816066741943, "identity": 0.508992612361908, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/ef625cc2201e1b923e143913.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "50040b28b0de45f7273775beed61a29e174209691615060cedd47a504dc045b8", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0639126300811768, "target_z": 1.6875, "ce": 6.268836498260498, "genuineness": 4.005263805389404, "identity": 0.2727321982383728, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/628c9ceda72677f80f5bf41a.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "bcd61832f9a730e8a6d452ff192478640a7908fbdd6fd6683f6b06b984abb9c8", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4397032260894775, "target_z": 0.9375, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.31985723972320557, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/01f2b5ddd69c9589db5c209a.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "d5f8525ea566a8ac9d914b1e81633f60ecdf4a5db4b347a48e238611dbc8f957", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7713146209716797, "target_z": 1.335937679040412, "ce": 6.173524379730225, "genuineness": 3.7291181087493896, "identity": 0.5432256460189819, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/a9bd863bb4ea74bd01ca5287.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "2c6a24b83dee2ce058244823fc9b7d528b5b3a73ef602c5b851afdab1da5a0c7", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1809518337249756, "target_z": 1.8281249283838352, "ce": 6.229123115539551, "genuineness": 3.7918784618377686, "identity": 0.45719113945961, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/9c68e0178891de93bfc29eda.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "bf719bf6cf87abc3e1bf5de1b8da6a635ca7340e97d76d9febae904b43138d46", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.491720676422119, "target_z": 1.0, "ce": 6.157639026641846, "genuineness": 3.9299514293670654, "identity": 0.46625372767448425, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/7085bb8e255908fe4db4177f.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "c7207ea9caee7f70e71f51759b83c06ebee63fada019ecfbc537db90811567b7", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7323012351989746, "target_z": 1.289062320959588, "ce": 6.086154937744141, "genuineness": 3.7542223930358887, "identity": 0.4864767789840698, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/29335e322411ebe5cc8696b4.mp3" }, { "key": "original-vn_TENS_reg_high-en-0-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "a78a21a429914b84e84ffd46a13b9b9c853ccef99cc9b7b72af211faa67f963e", "task_id": "vn_TENS_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.310995578765869, "target_z": 1.9843750716161648, "ce": 6.1020402908325195, "genuineness": 3.6412534713745117, "identity": 0.5983755588531494, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/b939306ab941d9bac662ad80.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "c598cc5067e05977cefc767a439beda56154f0d7e067b829002dce10ea92643a", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0625767707824707, "target_z": 0.48437507161616483, "ce": 6.173524379730225, "genuineness": 3.9801597595214844, "identity": 0.5849159359931946, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/b16d36653117dfb40c559e3c.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "8ff956577464c5bb33210e7ab3245476afcb0bb85bc0111f1398f15445c5f799", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2576422691345215, "target_z": 0.7187501432323297, "ce": 6.213237762451172, "genuineness": 4.111956596374512, "identity": 0.6116961240768433, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/06b699cb6fe497d4ae41de3c.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "ed5130f10ecda17fa5d6c9eea9745edab81a2761730a8a1931f40cf6a3a193fc", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0950875282287598, "target_z": 0.5234373925757527, "ce": 6.229123115539551, "genuineness": 3.9048471450805664, "identity": 0.648823618888855, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/3b2b731e22b6d3c4b6b2c9ab.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "d862c26e13832fb227175afedbf2801cbf103686e170949278eb1caee04db16b", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.6324167251586914, "target_z": -0.032470769146151976, "ce": 6.3641486167907715, "genuineness": 4.105680465698242, "identity": 0.5257452726364136, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/4ec0c9fa9125b196298787b6.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "a2fcfb4b0515969d4df678910e99cbb8a6b706b5caeced58c406893ebab22c0a", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.6326199769973755, "target_z": -0.032226558023989695, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.4141271114349365, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/en/39dae7f91b363163ba86d4fa.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "474f062d97deaeb22b56ecb96bb07cdbfd68212b44b46663e2c1d7df845910b7", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.130849599838257, "target_z": 0.5664062320959588, "ce": 6.332377910614014, "genuineness": 4.13078498840332, "identity": 0.5338538885116577, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/697db02357b23ebf4c425be7.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "c17dbae043253cadd15f7de85a77a1d3681ebf4829353d4649c09dc0ec04530b", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.530733823776245, "target_z": 1.0468750716161648, "ce": 6.292664527893066, "genuineness": 4.293961524963379, "identity": 0.2852476239204407, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/a7a90adf5e3e3e93764d1a6c.mp3" }, { "key": "original-vn_TENS_reg_high-en-1-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "18b592053fc29cb416d7e44d3082d2d6d7ae5b7735fc22c54623859173e52a9a", "task_id": "vn_TENS_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.6013282537460327, "target_z": -0.06982418517992273, "ce": 6.292664527893066, "genuineness": 4.2751336097717285, "identity": 0.5177371501922607, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/en/7ea73077d324e05a4f16f604.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "aa09a22d193cb99dc22c490984d25334c79e3bad7720e97a6f96e799cc0cdbff", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2576422691345215, "target_z": 0.7187501432323297, "ce": 6.070269584655762, "genuineness": 4.607763767242432, "identity": 0.6512030363082886, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/e1775b6d0ef5574502ee03a4.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "243c02b65ee77d7018310525c7e6234f2528c8c05d836603409114c46f0c1ef5", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.933868885040283, "target_z": 1.5312498567676702, "ce": 6.1894097328186035, "genuineness": 4.149612903594971, "identity": 0.569659948348999, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/44053dee22d66efbfdbea799.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "7f8b85b05e868c253a77ba2d8babf5b505fb368b30c295baf5862eac6672fcb4", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6347687244415283, "target_z": 1.1718750716161648, "ce": 6.165581703186035, "genuineness": 4.193545341491699, "identity": 0.6414066553115845, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/ca2597b762c7625d2e38d9ce.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "bceb77da6da99217dae8e94ac736d9fd7a5540a2cd7494d46872303621ae5171", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.34867262840271, "target_z": 0.8281249283838351, "ce": 5.911416053771973, "genuineness": 4.319066047668457, "identity": 0.5148633718490601, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/es/f8fa91d693a3af708bd870c1.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "8e108ba7e9d8ff259b434b8abf49bdeeb871b76b1d7aeacc9b1ade500f77a75b", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7062926292419434, "target_z": 1.2578124641919175, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.35392069816589355, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/es/13bbd239b1626875a9c4e241.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "9e08faba9caa511e2161a962149bfc51918beded0cb3b8d40f38ae2e246446e5", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3226640224456787, "target_z": 0.7968750716161649, "ce": 5.863759994506836, "genuineness": 4.149612903594971, "identity": 0.5033974051475525, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/62aae1c87c658028d7551cda.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "0deef604ef1066043daff15420c36b34df422fef10b70c3403a13d15ab4a0db3", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3356683254241943, "target_z": 0.8125, "ce": 5.935244083404541, "genuineness": 4.218649387359619, "identity": 0.6168591380119324, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/es/7bd7d7612e291caf4ed225f4.mp3" }, { "key": "original-vn_TENS_reg_high-es-0-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "7335f0a10a200a5931b3da0731e4b9b9a2c6cc75f6cb3709eec98f93198cab20", "task_id": "vn_TENS_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.091836452484131, "target_z": 0.519531160479794, "ce": 5.95112943649292, "genuineness": 4.268857479095459, "identity": 0.41595226526260376, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/fe3fd8c85f2e73ae05270a60.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "0caf54fe4902253a1cb12df7008d61e89b27ce0b54689b313d6a8a493146a5f5", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1503562927246094, "target_z": 0.5898439111363709, "ce": 6.348263263702393, "genuineness": 4.507347106933594, "identity": 0.6557530760765076, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/542d21d6219eb1e4a5f7c238.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "cc672a78ed2e930b03ee6738e8d15c064b28bd67214afc556c91d53ba7f8e363", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.8691368103027344, "target_z": 0.2519531160479794, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.510194718837738, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/28f79e72fbf8da7e173bc5f7.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "dac845b267c93276913d9291f8a8607b6328b3c9d5895be07c0480be60ff6dd8", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.8041150569915771, "target_z": 0.17382818766414426, "ce": 6.340320587158203, "genuineness": 4.394378662109375, "identity": 0.5946924686431885, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/f7820ccbd98da5d650113a2a.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "36825fbeef7ed46dffe8cf93600c9bb2664e2f83a1f3d629804e8cca0aa2c1b5", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.8398770093917847, "target_z": 0.2167968839520206, "ce": 6.252951145172119, "genuineness": 4.507347106933594, "identity": 0.6651703119277954, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/584b42a4f88ef26dfd770357.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "e04aa882cea47601bc477a0681fbd9c5d6085fe0b39442100d055ae27c909a53", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.608759880065918, "target_z": 1.1406249283838352, "ce": 5.871703147888184, "genuineness": 4.482243061065674, "identity": 0.667220950126648, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/412a79d9588df9f4f5a8e117.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "b4c1efbc8f032900a177dc1962a77474d69d2cdb35f933ba94bf1bab67ebbf06", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.9081499576568604, "target_z": 0.29882818766414426, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.6634107828140259, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/70b39e931a7fad42f5923808.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "f325335f2e0a7b03fe237a9705ee8b915b6da669f82cf357a0a940ae4b92d752", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4071922302246094, "target_z": 0.8984373925757527, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.5762217044830322, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/es/cda519403008d6c30613bc3b.mp3" }, { "key": "original-vn_TENS_reg_high-es-1-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "340de3390a2fb6dcb639310a64429e79aa1fb364e8ed7bf0d77c96be22760def", "task_id": "vn_TENS_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0365679264068604, "target_z": 0.45312492838383517, "ce": 5.982900142669678, "genuineness": 4.3881025314331055, "identity": 0.4700915813446045, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_TENS_reg_high/es/914eca28865a1b2b5549b124.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "3b1e8a7a7266280befdd36a79a59aab8c372cd18c225e8860ba7c3397398ddc9", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1373517513275146, "target_z": 0.5742186962878764, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.8295826315879822, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/7576848169ccb53a0748d0d1.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "e75ad7f348bb60d5dc05943f3866f526d12c84edfae9a86e74c2cd54c0ac2723", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1113431453704834, "target_z": 0.542968839520206, "ce": 6.268836498260498, "genuineness": 4.595211505889893, "identity": 0.8214932680130005, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/e7beec7ef83fce25d789f7ef.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "d3ae765b021d8cb37e77cf23b308b0d4bf9047c505801d904590e84d34a8996c", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.9081499576568604, "target_z": 0.29882818766414426, "ce": 5.895530700683594, "genuineness": 4.042920112609863, "identity": 0.6214881539344788, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/2a5e93d6703b64eab4ba7252.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "72f78930b3221ecde8e3779c1b6ab85d71a6bfd65b2ac24b8b486fb3e08aa26e", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972882032394409, "target_z": 1.5781249283838352, "ce": 6.252951145172119, "genuineness": 4.105680465698242, "identity": 0.7341771721839905, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/f74822ae1d1a5d2d07256637.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "0eff795044e796a3e9e710221bd04f00910cbb76773e483553bb56815c0f6a6d", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.238135576248169, "target_z": 0.6953124641919176, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.7124825716018677, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/9d3cd434fdf957005634bf6f.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "ad7572ee982a548450870097e325a83e5558a807ed2386275966c769108e7c7e", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.9130265712738037, "target_z": 0.30468753580808244, "ce": 6.221180438995361, "genuineness": 4.369274139404297, "identity": 0.7505081295967102, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/ed3675faabf3e0bfd0546c22.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "304efb6cb0b68d15eafe9a7bd54ec97ccb07d773585e7b349a56a3a065f263f7", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.351923704147339, "target_z": 0.832031160479794, "ce": 6.054384231567383, "genuineness": 4.394378662109375, "identity": 0.5292630791664124, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/66ef0c7150a18597b9dce525.mp3" }, { "key": "original-vn_TENS_reg_high-fr-0-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "54837eb2f90e4660905e8994198894ff6794a327afe41c39a22f07cd96fb1e19", "task_id": "vn_TENS_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.9130265712738037, "target_z": 0.30468753580808244, "ce": 6.1020402908325195, "genuineness": 4.557555198669434, "identity": 0.6690945625305176, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/19215f82badd7e26a51911b1.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "d8c8901d6f7e366201ac3d0627dbd4e5fcb89bda7e2f93dd56a2a59c5c45969b", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.7061759233474731, "target_z": 0.056152372844066975, "ce": 6.070269584655762, "genuineness": 4.494794845581055, "identity": 0.7799606323242188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/2e9bc82904dc9a75282d1518.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "c27076678065f6480d939c1d4008a812f854d18d1206f21b252a8adc037a5d1c", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2121267318725586, "target_z": 0.6640623209595878, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.7400606870651245, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/8e14abc8b52833e035568cb5.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "66f749b1d1af207069b96576734f21e3856fa15a66c0cd901708caf214c2bd54", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5957555770874023, "target_z": 1.125, "ce": 6.1894097328186035, "genuineness": 4.300237655639648, "identity": 0.7617445588111877, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/205321608e928e69a8167eaf.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "04f97fb802261746ce39895d5b9424a87e9859356f85c59e548cb5f0a7175a59", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195871353149414, "target_z": 0.644531160479794, "ce": 6.125868320465088, "genuineness": 4.306513786315918, "identity": 0.7037789225578308, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/bb0df30898a92095264f1c40.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "a9f85b11e19f66bcf90b86373b37a4d0844b3d68682651567d5890cd335ee4d0", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.7090206146240234, "target_z": 0.05957032592803091, "ce": 6.0146708488464355, "genuineness": 4.168440818786621, "identity": 0.6800476908683777, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/3c387cfa75a5286ea101519b.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "5fa7467aab55d50d2263a3b5f75ebdae1f41c61c3a970ae363ffa1550d5d532a", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1373517513275146, "target_z": 0.5742186962878764, "ce": 6.173524379730225, "genuineness": 4.494794845581055, "identity": 0.74281907081604, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/c8fc6f98ec2a15541188da89.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "ba8c134b10380e4627c15cfe5b4eeaed69a4a8b9d27cb37203583720303777ae", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.93740975856781, "target_z": 0.333984419760103, "ce": 6.09409761428833, "genuineness": 4.3504462242126465, "identity": 0.7705447673797607, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/c861b587ce1cd050c94ffae2.mp3" }, { "key": "original-vn_TENS_reg_high-fr-1-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "8581aeef51d38d9c685b7a457f3cb93be72f66e5039917ebcef3abd36683bc56", "task_id": "vn_TENS_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 1.86751127243042, "target_z": 0.25, "ce": 6.125868320465088, "genuineness": 4.507347106933594, "identity": 0.7336236238479614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/fr/49ee11d4d720c536a76f7119.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "9e5831e6a84dbe0b8e12287ab33a0868de1609b179d0233e979e586bc9efa819", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.933868885040283, "target_z": 1.5312498567676702, "ce": 6.23706579208374, "genuineness": 4.036643981933594, "identity": 0.5096359848976135, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/7019305720afad3180d53731.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "2d5f54e0ba8dbc1622292d8334565c6799f5d394da469b5ba1b1a172c2f808f7", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.044405937194824, "target_z": 1.664062320959588, "ce": 6.181467056274414, "genuineness": 4.394378662109375, "identity": 0.5217533707618713, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/1b7f177f214d3577a67de7ec.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "ab082277e102137a38eacf7b7ebca681d6ba82869cb9e1f8afd4b26df7abb612", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1809518337249756, "target_z": 1.8281249283838352, "ce": 6.1020402908325195, "genuineness": 3.9550557136535645, "identity": 0.4906318783760071, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/8bdac206d5cb6169f4b5396f.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "28ce46e9185207b9d28bf29a12b64b1818f1c334d065034e8a4ee9eed8eb259f", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2264671325683594, "target_z": 1.8828124641919175, "ce": 6.117925643920898, "genuineness": 4.105680465698242, "identity": 0.5826599597930908, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/c167e81db1ef3d981390d51d.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "17b5c718a591b2a713028acf48334186564f69aa3ae423065c83892d2fef96fb", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.68678617477417, "target_z": 1.2343750716161648, "ce": 6.022613525390625, "genuineness": 4.20609712600708, "identity": 0.39960968494415283, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/7b26b326bf247d5c335640b3.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "cc5ef78724713feac502cd24a83ab1313002d5c3162ec9f5f0ddf8b22d2d5403", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.16794753074646, "target_z": 1.8125, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.5300474166870117, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/4a73777b60f89147796a81fa.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "91c3e9ee2f7051a6952ff2fb76a8f219d71541a0a5a12763fee2806696d7b7f0", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532069683074951, "target_z": 2.25, "ce": 5.982900142669678, "genuineness": 3.9048471450805664, "identity": 0.7334014177322388, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/36ac4473c831d3c962258fdc.mp3" }, { "key": "original-vn_TENS_reg_high-nl-0-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "17047a01f0ebb88da927b9d801900f834ba0f6cb8cde2baf8b59cdd8cf859c74", "task_id": "vn_TENS_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.128934383392334, "target_z": 1.7656249283838352, "ce": 6.205295085906982, "genuineness": 4.143336772918701, "identity": 0.8176337480545044, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/b888ba1e8437e7d66d711b27.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1701-scale0.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "232e7d402b0a6a9c1f7d01d0048659f5a5d5c95e2befc073a2b30c6dde009412", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.050908327102661, "target_z": 1.6718750716161648, "ce": 6.173524379730225, "genuineness": 4.055472373962402, "identity": 0.6960249543190002, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/376f3802d4ecee295dcf8f31.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1701-scale0.5", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "6e1e4e915120a340700c48bae78f3363dfdcd6974dbac93aee2d172c29e65513", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0834193229675293, "target_z": 1.710937679040412, "ce": 6.268836498260498, "genuineness": 4.300237655639648, "identity": 0.6312883496284485, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/5b7676fa4e6431b5b555db77.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1701-scale0.75", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "86edf454ec81b9734e2db4988f947db5e908e69a6b5e37ab7e0a51186fd7e655", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545073986053467, "target_z": 2.265624928383835, "ce": 6.221180438995361, "genuineness": 3.9299514293670654, "identity": 0.5699334144592285, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/aeed73de5a96b9d1d64a30c6.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1701-scale1.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "0c662a343e7cdcc11706cfd4c7ebc2d75a43100826e1d64043b8029af7db9869", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0834193229675293, "target_z": 1.710937679040412, "ce": 6.181467056274414, "genuineness": 3.9801597595214844, "identity": 0.7022305130958557, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/cf732b2b2e3e0b0fc7c51183.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1702-scale0.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "4e32a8901df6d81399ddb39cb83f055f94dbe6d86c8a6473915fdbc6fbd624b0", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545073986053467, "target_z": 2.265624928383835, "ce": 6.24500846862793, "genuineness": 3.8671910762786865, "identity": 0.7765212059020996, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/3ddee95c48882a8e5ed0ada8.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1702-scale0.5", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "9adb7607930b63815b5185d10361476607073d136d97e88c99df782234a3412c", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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": 2.9208645820617676, "target_z": 1.5156249283838352, "ce": 6.292664527893066, "genuineness": 4.394378662109375, "identity": 0.6332422494888306, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/51ed775eddf10bf2ee5c24e9.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1702-scale0.75", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "78201956ddead42d1224609e4dd9c8d40fb42d63e328db1f8c7a9a1bace6ad77", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4800522327423096, "target_z": 2.1875, "ce": 6.022613525390625, "genuineness": 3.9173994064331055, "identity": 0.45768919587135315, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/e3265c60a62f1cd865d7de02.mp3" }, { "key": "original-vn_TENS_reg_high-nl-1-seed1702-scale1.0", "bucket": "vn_TENS_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "1094cbf0a1d43f2ef8d0cfd660b4fae70566e137913c45e184e79bbb769b8d7a", "mp3_sha256": "a64691b415c138b32a961e29413c4d3c5f0fc4f0025c7e10244b28dc0cba3a6b", "task_id": "vn_TENS_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very tense and strained. 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 tense and strained. 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 tense and strained. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.350008487701416, "target_z": 2.0312498567676704, "ce": 5.780362129211426, "genuineness": 3.9550557136535645, "identity": 0.48424267768859863, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het Oude Raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_high/nl/bb1c3b1d5f4f66760676cdcd.mp3" } ], "source_standard_audio_unchanged": true }