{ "complete": true, "report": { "bucket": "vn_S_MONO_reg", "label": "Strong introspective monologue delivery, not monotone speech", "definition": { "id": "vn_S_MONO_reg", "category": "talking_style", "head": "vn_S_MONO_reg", "pole": "high", "label": "Strong introspective monologue delivery, not monotone speech", "floor": 4.0, "canonical_dimension_name": "Monologue Style" }, "recipe_status": "corrected_taxonomy_new_search", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P02", "general": "An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "inline": [ "An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.", "An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 5.0 }, "n": 20, "target_z": 1.2025390302740586, "target_raw": 4.196071445941925, "ce": 6.095686221122742, "genuineness": 4.5346479415893555, "identity": 0.5876939356327057, "wer": 0.021246775025799792, "valid_fraction": 1, "identity_delta": 0.03841220438480377, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.3365234092634144, "target_raw": 4.366541481018066, "ce": 5.991935062408447, "genuineness": 4.434231233596802, "identity": 0.6065645143389702, "wer": 0.05182469040247678, "valid_fraction": 0.9, "identity_delta": 0.05728278309106827, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.2025390302740586, "target_raw": 4.196071445941925, "ce": 6.095686221122742, "genuineness": 4.5346479415893555, "identity": 0.5876939356327057, "wer": 0.021246775025799792, "valid_fraction": 1, "identity_delta": 0.03841220438480377, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.131835934883129, "target_raw": 4.106114995479584, "ce": 6.085757851600647, "genuineness": 4.438938224315644, "identity": 0.6165027558803559, "wer": 0.024385964912280702, "valid_fraction": 0.95, "identity_delta": 0.06722102463245391, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 0.8449707029923265, "target_raw": 3.7411327004432677, "ce": 6.1516820907592775, "genuineness": 4.39939923286438, "identity": 0.6460829526185989, "wer": 0.057653508771929825, "valid_fraction": 0.95, "identity_delta": 0.09680122137069702, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.8103515594073344, "target_raw": 3.6970863223075865, "ce": 6.171935796737671, "genuineness": 4.390612721443176, "identity": 0.5946034997701645, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04532176852226257, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.7273437489020123, "target_raw": 3.5914744257926943, "ce": 6.177495741844178, "genuineness": 4.391554141044617, "identity": 0.5578406497836113, "wer": 0.01215686274509804, "valid_fraction": 0.95, "identity_delta": 0.008558918535709382, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 0.6173339886983269, "target_raw": 3.451507604122162, "ce": 6.179878520965576, "genuineness": 4.3366388320922855, "identity": 0.5262946583330631, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.022987072914838792, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 0.6160888764128848, "target_raw": 3.4499234318733216, "ce": 6.190005445480347, "genuineness": 4.347308123111725, "identity": 0.5843401998281479, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.03505846858024597, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 0.5518066495873253, "target_raw": 3.3681363344192503, "ce": 6.117925691604614, "genuineness": 4.271995627880097, "identity": 0.5722048684954644, "wer": 0.021900584795321636, "valid_fraction": 0.95, "identity_delta": 0.022923137247562408, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.3365234092634144, "target_raw": 4.366541481018066, "ce": 5.991935062408447, "genuineness": 4.434231233596802, "identity": 0.6065645143389702, "wer": 0.05182469040247678, "valid_fraction": 0.9, "identity_delta": 0.05728278309106827, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.3140625449442997, "target_raw": 4.337964236736298, "ce": 5.958972978591919, "genuineness": 4.400026822090149, "identity": 0.5991055130958557, "wer": 0.03720287237702098, "valid_fraction": 0.9, "identity_delta": 0.049823781847953795, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.2673828178252406, "target_raw": 4.2785730242729185, "ce": 5.998611831665039, "genuineness": 4.3730398416519165, "identity": 0.562417495250702, "wer": 0.0746813725490196, "valid_fraction": 0.9, "identity_delta": 0.013135764002799987, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.1843750541673972, "target_raw": 4.172961187362671, "ce": 6.045250248908997, "genuineness": 4.385278129577637, "identity": 0.5885573908686638, "wer": 0.006274509803921568, "valid_fraction": 1, "identity_delta": 0.03927565962076187, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.1439453334898664, "target_raw": 4.121521937847137, "ce": 6.141753697395325, "genuineness": 4.384964394569397, "identity": 0.6508373111486435, "wer": 0.03341374269005848, "valid_fraction": 0.95, "identity_delta": 0.10155557990074157, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.1363281114947503, "target_raw": 4.111830449104309, "ce": 6.081587958335876, "genuineness": 4.387474799156189, "identity": 0.62164938300848, "wer": 0.02103844169246646, "valid_fraction": 1, "identity_delta": 0.07236765176057816, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.0319336306149396, "target_raw": 3.9790080189704895, "ce": 6.154660534858704, "genuineness": 4.402537274360657, "identity": 0.5585081808269023, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": 0.009226449579000474, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.0144531311121319, "target_raw": 3.9567673563957215, "ce": 6.07682237625122, "genuineness": 4.433917427062989, "identity": 0.5637034378945828, "wer": 0.015465686274509803, "valid_fraction": 1, "identity_delta": 0.014421706646680832, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.9636718697662583, "target_raw": 3.8921577095985413, "ce": 6.128846883773804, "genuineness": 4.42136538028717, "identity": 0.569375665858388, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.02009393461048603, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 0.9435546629965728, "target_raw": 3.8665623307228087, "ce": 6.052001595497131, "genuineness": 4.412578868865967, "identity": 0.6205041080713272, "wer": 0.015302287581699346, "valid_fraction": 1, "identity_delta": 0.07122237682342529, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 0.9270507928795205, "target_raw": 3.8455642461776733, "ce": 6.0956861734390255, "genuineness": 4.338207817077636, "identity": 0.5598950013518333, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.010613270103931427, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.8396484452774134, "target_raw": 3.7343611240386965, "ce": 6.093700528144836, "genuineness": 4.3184382796287535, "identity": 0.5827678620815278, "wer": 0.01515608875128999, "valid_fraction": 1, "identity_delta": 0.033486130833625796, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.804687511858268, "target_raw": 3.6898798823356627, "ce": 6.188615489006042, "genuineness": 4.406302857398987, "identity": 0.6170336499810218, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": 0.06775191873311996, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.7788574049888629, "target_raw": 3.6570159077644346, "ce": 6.146320724487305, "genuineness": 4.380571079254151, "identity": 0.5876068606972694, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03832512944936752, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 0.7558105536687993, "target_raw": 3.6276931047439573, "ce": 6.172730088233948, "genuineness": 4.294275569915771, "identity": 0.573260585218668, "wer": 0.0059649122807017545, "valid_fraction": 1, "identity_delta": 0.023978853970766066, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 0.7527344076834362, "target_raw": 3.623779284954071, "ce": 6.185438394546509, "genuineness": 4.307141387462616, "identity": 0.598572951555252, "wer": 0.012423245614035088, "valid_fraction": 1, "identity_delta": 0.049291220307350156, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.7268554951108566, "target_raw": 3.590853214263916, "ce": 6.13937087059021, "genuineness": 4.336638808250427, "identity": 0.5570714019238949, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": 0.007789670675992966, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 0.7196777586817045, "target_raw": 3.5817208886146545, "ce": 6.160816097259522, "genuineness": 4.327852344512939, "identity": 0.5672520369291305, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.01797030568122864, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 0.6954101651345511, "target_raw": 3.5508449196815492, "ce": 6.210457801818848, "genuineness": 4.334128475189209, "identity": 0.5552211046218872, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.00593937337398529, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 0.6832030989044905, "target_raw": 3.5353137135505674, "ce": 6.15108630657196, "genuineness": 4.332245612144471, "identity": 0.5487686924636364, "wer": 0.00804093567251462, "valid_fraction": 1, "identity_delta": -0.0005130387842655181, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 0.6614990181236966, "target_raw": 3.5076993346214294, "ce": 6.156249046325684, "genuineness": 4.3287938117980955, "identity": 0.5611519854515791, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.011870254203677178, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 0.6214843567368032, "target_raw": 3.456788170337677, "ce": 6.185438394546509, "genuineness": 4.2754474639892575, "identity": 0.5841557428240776, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03487401157617569, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 0.5932128965861586, "target_raw": 3.420818030834198, "ce": 6.2025151491165165, "genuineness": 4.392495584487915, "identity": 0.5833778247237206, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.034096093475818635, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 0.5914062797920684, "target_raw": 3.418519449234009, "ce": 6.154461979866028, "genuineness": 4.312789869308472, "identity": 0.5704583019018173, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.021176570653915407, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 0.5831054687591499, "target_raw": 3.407958221435547, "ce": 6.172332978248596, "genuineness": 4.3065138220787045, "identity": 0.5492817312479019, "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": 0.6643066343618947, "target_raw": 3.5112715005874633, "ce": 6.146122145652771, "genuineness": 4.271995663642883, "identity": 0.6021968096494674, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.05291507840156555, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 0.6009765544297898, "target_raw": 3.4306958317756653, "ce": 6.201323771476746, "genuineness": 4.292706501483917, "identity": 0.5908856496214867, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.04160391837358475, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 0.6520751980734818, "target_raw": 3.4957092881202696, "ce": 6.185438418388367, "genuineness": 4.327852332592011, "identity": 0.5474852129817009, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.0017965182662010193, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 0.4482177785619005, "target_raw": 3.2363388919830323, "ce": 6.212602329254151, "genuineness": 4.331115899085998, "identity": 0.6053401204943657, "wer": 0.009232072275550536, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.0026562520349374, "target_raw": 3.9417580366134644, "ce": 6.221180410385132, "genuineness": 4.490401792526245, "identity": 0.6051376783847808, "wer": 0.008488612836438923, "valid_fraction": 1, "identity_delta": -0.00020244210958480835, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.0978125094573348, "target_raw": 4.062826557159424, "ce": 6.143421659469604, "genuineness": 4.416093521118164, "identity": 0.6289109975099564, "wer": 0.009318181818181817, "valid_fraction": 1, "identity_delta": 0.023570877015590668, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.0026562520349374, "target_raw": 3.9417580366134644, "ce": 6.221180410385132, "genuineness": 4.490401792526245, "identity": 0.6051376783847808, "wer": 0.008488612836438923, "valid_fraction": 1, "identity_delta": -0.00020244210958480835, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.09515625742239739, "ci95_z": [ 0.05718748913724092, 0.13312502570755386 ], "language_deltas": { "en": 0.04140623762933772, "de": 0.12578130116623049, "fr": 0.04726562042505095, "nl": 0.15273439831394053, "es": 0.10859372957742729 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.07775875091552735, "delta_genuineness": -0.07430827140808105, "delta_wer": 0.0008295689817428949, "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": 0.09515625742239739, "ci95_z": [ 0.05718748913724092, 0.13312502570755386 ], "language_deltas": { "en": 0.04140623762933772, "de": 0.12578130116623049, "fr": 0.04726562042505095, "nl": 0.15273439831394053, "es": 0.10859372957742729 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.07775875091552735, "delta_genuineness": -0.07430827140808105, "delta_wer": 0.0008295689817428949, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.5544384734730369, "ci95_z": [ 0.4486914103180447, 0.642998032499595 ], "language_deltas": { "en": 0.6002929653919874, "de": 0.5143554447498171, "fr": 0.6840332268552608, "nl": 0.6233153716977329, "es": 0.3501953586703861 }, "languages_improved": 5, "clear_gain": true, "delta_ce": 0.008578081130981446, "delta_genuineness": 0.15928589344024657, "delta_wer": -0.0007434594391116131, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": true, "scale": 1.0, "cfg": 5.0, "reason": "Held-out target improves against matched CFG and best manual-only configuration; QA passes" }, "reference_only": true, "updated": 1793035260.2788713, "example": { "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "language": "nl", "seed": 1701, "scale": 1.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "000f19a645a36e304df770b858fae74e28b536909d63d40feee79b5dd924e2f6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "cb09726c9f74319c005bcd0bae736d4664364cda0c381b5fd93c8ee45c1fb944", "generation_seconds_batch": 24.70726302091498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 5.9431867599487305, "genuineness": 4.268857479095459, "identity": 0.772404134273529, "hypothesis": "La porte s'est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "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_S_MONO_reg/000f19a645a36e304df770b858fae74e28b536909d63d40feee79b5dd924e2f6.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0023dcaf648ace09fa215472e96042166ef8f505433bd5d1aa359fc46a86fbde", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7c704f6581bf5018b5c8fd1fae97631f27084156ab0b26c03f8ecde7a24d9d56", "generation_seconds_batch": 30.363590634893626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.289797067642212, "target_z": 0.49023431515966603, "ce": 5.927301406860352, "genuineness": 4.042920112609863, "identity": 0.7246457934379578, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0023dcaf648ace09fa215472e96042166ef8f505433bd5d1aa359fc46a86fbde.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "018254b86862ad5dcdc83ad505f8ea7e4463cd2dfdee8cadb18bceef6c5544b9", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "02a9b3490375624c9c8ed1f4996f7c125cfb39533b012a47f1e6c34b13effcd0", "generation_seconds_batch": 17.133620452019386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.62030029296875, "target_z": 0.7500000234237393, "ce": 6.229123115539551, "genuineness": 4.143336772918701, "identity": 0.6793301701545715, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/018254b86862ad5dcdc83ad505f8ea7e4463cd2dfdee8cadb18bceef6c5544b9.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "019be372257e95e12b2bf970c654c2204d7e30f5a999a5de125a3f73616104bc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d68b56438622950eb02b2569b6fe7f64bb9e9a126225600e3dd4012133d04ae8", "generation_seconds_batch": 18.73556381301023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.6582269668579102, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/019be372257e95e12b2bf970c654c2204d7e30f5a999a5de125a3f73616104bc.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "01d72ffd82ef4e9d312ee9169b939bf27338db8ed9ac98c82581c17e9d1a8a41", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0f183758a9b3c5c0ab07f1dd0bba2b5320f709d49dc83428d418bd6ef67a5be1", "generation_seconds_batch": 17.332801937009208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0412986278533936, "target_z": 0.29492193008238693, "ce": 6.3641486167907715, "genuineness": 4.6203155517578125, "identity": 0.5786575078964233, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/01d72ffd82ef4e9d312ee9169b939bf27338db8ed9ac98c82581c17e9d1a8a41.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "01e41a7ef3226b5857b0191cf103d1272b054a92f82759a6e5dc323d390be34e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "aedd4e0410d8c86688c4c92f6b06e657627116024ec1d6419da79aaf2b444031", "generation_seconds_batch": 24.389690459007397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.173524379730225, "genuineness": 4.262581825256348, "identity": 0.4918786287307739, "hypothesis": "ik heb het kleine doosje gevonden, het stond naast het oude Araam. 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_S_MONO_reg/01e41a7ef3226b5857b0191cf103d1272b054a92f82759a6e5dc323d390be34e.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "db62afd6aac6355a8b79a964360ddab8bcfcf024fce252afdc80c83d7d2997a5", "generation_seconds_batch": 15.887489561922848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.7954570055007935, "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_S_MONO_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "024d3a0336dc11e4dc79d09ea1f5b1dd271118c18708d7c587dcfe67f32e3208", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "54bf17b7f6c68ca0197b78bb3d958cc07c245a048853d54ab6dfc0a3b1e55c31", "generation_seconds_batch": 17.848345541977324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0239036083221436, "target_z": 0.28124996193642365, "ce": 6.221180438995361, "genuineness": 4.557555198669434, "identity": 0.596900224685669, "hypothesis": "La porte est encore rouverte. 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_S_MONO_reg/024d3a0336dc11e4dc79d09ea1f5b1dd271118c18708d7c587dcfe67f32e3208.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "0293f6123095a2fcbcb4fe0e7830fcaffba9bf32318bbf7f9bd2b03ba1a6a647", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d6055db4e37cf07095f38f56953fba98d00e3cbdc39f00aefbde8500396f033e", "generation_seconds_batch": 24.520888195023872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.38277414441108704, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0293f6123095a2fcbcb4fe0e7830fcaffba9bf32318bbf7f9bd2b03ba1a6a647.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "02e9cd1ab0b8cb26f59a2afd04ffd724be51619b5faf61b0ae12511baaea8f2c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5f7276ead45f2cd0dd77be9cb3260948d3f1b05041a1cf21d38d520462248c74", "generation_seconds_batch": 24.388625669060275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.2995469570159912, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/02e9cd1ab0b8cb26f59a2afd04ffd724be51619b5faf61b0ae12511baaea8f2c.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "034aaed57f864bb8ee90e290fcce162f9192983c114937cdb129a3ebd60e07ed", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c7339893ff69122ec1ae1d4c2d2f3bf0fe456dfd3ffa507104916d25a4da19f3", "generation_seconds_batch": 24.46639031090308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545750617980957, "target_z": 0.691406195466607, "ce": 6.300607204437256, "genuineness": 4.218649387359619, "identity": 0.3184545040130615, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/034aaed57f864bb8ee90e290fcce162f9192983c114937cdb129a3ebd60e07ed.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "0396492b313bfb61db87156d509aa9978e5d3397d401c3389dfff15046ab276f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7e39dc4dc66cc756c5e1d28590fa4a55a249941a23d7fd4c6fef3b561d9648d9", "generation_seconds_batch": 17.874014395056292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7843093872070312, "target_z": 0.878906295017499, "ce": 6.197352409362793, "genuineness": 4.137060642242432, "identity": 0.4233954846858978, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0396492b313bfb61db87156d509aa9978e5d3397d401c3389dfff15046ab276f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "041112b1357965386f7f5e02f376a1f45b6e19ecf39164fdf54634a5377fe918", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ee09573d59ac7ee3746567912d88c33b0e418315d52a1befed18988d91a4e6d7", "generation_seconds_batch": 21.590709432959557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.7111687064170837, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/041112b1357965386f7f5e02f376a1f45b6e19ecf39164fdf54634a5377fe918.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "045a94fffe53705efe0329091d5e7c459a488f592b56db904b7a92f064c764c2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a1f96f97232dc460775771851de4170ee46d0f63a32fa442234470e198b3080c", "generation_seconds_batch": 17.25613145006355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.272402286529541, "target_z": 0.47656253440361707, "ce": 6.316492557525635, "genuineness": 4.381826400756836, "identity": 0.4763385057449341, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/045a94fffe53705efe0329091d5e7c459a488f592b56db904b7a92f064c764c2.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "04a42cc5b870b2afedc6e8ff9133214f5bed931d5a4c2263de7bf4b7e68aefe5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6ef094243a3f81b06dbb8d757b4268be41fb0531d3a558a3f4e0bbed18c929f9", "generation_seconds_batch": 22.243789308005944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.375736236572266, "target_z": 1.343750213741621, "ce": 6.165581703186035, "genuineness": 4.432034492492676, "identity": 0.513560950756073, "hypothesis": "De deur 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_S_MONO_reg/04a42cc5b870b2afedc6e8ff9133214f5bed931d5a4c2263de7bf4b7e68aefe5.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "055e37617c2edba6134808dfb574f64439900b1ac7629a1badc555a2a4fa57d0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "02213bbe110ab4f64e4ba6acd6848cbe212962858fe1cf286b15892f51bb4a26", "generation_seconds_batch": 26.026475038030185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.202822685241699, "target_z": 0.42187503659959263, "ce": 6.117925643920898, "genuineness": 4.268857479095459, "identity": 0.669583261013031, "hypothesis": "De deur 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_S_MONO_reg/055e37617c2edba6134808dfb574f64439900b1ac7629a1badc555a2a4fa57d0.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "05887a2e093aa263bb6bc8ce89d8aa206ebd27f1a25bfb46f4a7733743109319", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ba43129f7c790dadd9cbf889f8966fe4d7a11a500f22869af138eace985a06d", "generation_seconds_batch": 22.968772301916033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4950151443481445, "target_z": 1.4374998887372383, "ce": 6.109982967376709, "genuineness": 4.921566009521484, "identity": 0.8061830997467041, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/05887a2e093aa263bb6bc8ce89d8aa206ebd27f1a25bfb46f4a7733743109319.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "058939377c26b817c62b7d0181de045fdee55d6ad1310d78305a6ab456708edc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d045fdf20c45a3eb83ab66d057f2be167fd3314402f31facb4c2347ce2ab7bea", "generation_seconds_batch": 23.75978673994541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.341981887817383, "target_z": 0.5312500322076416, "ce": 6.181467056274414, "genuineness": 4.281409740447998, "identity": 0.36762118339538574, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/058939377c26b817c62b7d0181de045fdee55d6ad1310d78305a6ab456708edc.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "c3a2316a8e68812556992e9298c1c9688aed3c5b7a37b50e9472cd5f83b513d2", "generation_seconds_batch": 21.142702881013975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9667489528656006, "target_z": 0.2363281021252546, "ce": 6.356205940246582, "genuineness": 4.570107460021973, "identity": 0.6661509275436401, "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_S_MONO_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "05a0cecb52bee044ccec1f216a3bdf144b1ced615612be6cc87391f0d827907f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2f1b6d005aa120323a4a8ac8d1cf5990dae3bc845870975b9df72d8b8a0e5a83", "generation_seconds_batch": 24.199242640985176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.37083369493484497, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/05a0cecb52bee044ccec1f216a3bdf144b1ced615612be6cc87391f0d827907f.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "05a60a0dad44e73cdf051fcb270168bf17e2b7630c53fb7a991ba2a8d0f53b17", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7e3d32e3e5a2e19da912de8eaca7f49ddeefee6599ea6481bf1d5d027bc8afa3", "generation_seconds_batch": 23.265318181016482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.2767791748046875, "genuineness": 4.2751336097717285, "identity": 0.34733647108078003, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/05a60a0dad44e73cdf051fcb270168bf17e2b7630c53fb7a991ba2a8d0f53b17.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "06aa6b3541d1a688e08ffb37f14b680bab6bef52760bb28f40490bde3a00e063", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0609509d6d9e0dfc77e4c4a09c2a8cf389cf18b0830953ad6cc5df5331e37907", "generation_seconds_batch": 17.21346308698412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.300607204437256, "genuineness": 4.331617832183838, "identity": 0.6608914732933044, "hypothesis": "La porte 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_S_MONO_reg/06aa6b3541d1a688e08ffb37f14b680bab6bef52760bb28f40490bde3a00e063.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "06f4fd61ac2d2efb9f9a18e92e9f03affaf92de3c0b216c8ed4bdd9900c29882", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9ebd5795402e8ee1c36526ae1170ad0c47a2f2b459737b7f8e51a59d7806d865", "generation_seconds_batch": 23.529472123016603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8439488410949707, "target_z": 0.9257811325153077, "ce": 6.427689552307129, "genuineness": 4.432034492492676, "identity": 0.5677049160003662, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/06f4fd61ac2d2efb9f9a18e92e9f03affaf92de3c0b216c8ed4bdd9900c29882.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "07211b169ab53ac2f7cad732aefb64a2a1380e8ebf1f4982a10a8bbeba7a864d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "862ab2ba721f9641c3418ccc51b4cc080f032541487533b40ea67d9de01a978d", "generation_seconds_batch": 21.988214793032967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.227672576904297, "target_z": 0.4414063125853034, "ce": 6.078212261199951, "genuineness": 4.287685871124268, "identity": 0.6817910075187683, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/07211b169ab53ac2f7cad732aefb64a2a1380e8ebf1f4982a10a8bbeba7a864d.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0833ede8d968c65b238e54dc469f4881c8edcf02cf4a8d153fac259359af65b9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8b995d5b6813c9bf2bc7ee3ccc664dbacb0c896f473108ffada97d5b1b05faec", "generation_seconds_batch": 25.644426582963206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.229123115539551, "genuineness": 4.425758361816406, "identity": 0.6678963899612427, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0833ede8d968c65b238e54dc469f4881c8edcf02cf4a8d153fac259359af65b9.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "08f9336da303348d5bbe19488f2f420e5c7941ee15cc4235ad054ef095a15d2c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4276ff95cb51b48f0d70d570459657e12604933989a4f114579770522678d98c", "generation_seconds_batch": 17.874014395056292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.27948564291000366, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/08f9336da303348d5bbe19488f2f420e5c7941ee15cc4235ad054ef095a15d2c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "0981f88f480f934157ab8478dc48b68b5f87f556c9356cf9855a5c4eb3073f09", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "82aca29730e780fb4dab1332ba3d85fea1ccd579474275a5f30c064653a8db4d", "generation_seconds_batch": 24.331913403002545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.062326908111572, "genuineness": 4.231201171875, "identity": 0.5638411045074463, "hypothesis": "Die Tür 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_S_MONO_reg/0981f88f480f934157ab8478dc48b68b5f87f556c9356cf9855a5c4eb3073f09.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "09b08c097876406a8d402df84e52bbb23340b251ced78d22ade3ddbce7a9839d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "285bf3f50b18023b2e69181e04c0cc849713d398fdbf501c5c7da4698f4e518c", "generation_seconds_batch": 17.376624096999876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.504955291748047, "target_z": 1.4453125490434542, "ce": 5.700935363769531, "genuineness": 4.545003414154053, "identity": 0.5337258577346802, "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_S_MONO_reg/09b08c097876406a8d402df84e52bbb23340b251ced78d22ade3ddbce7a9839d.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "09bdba9c1fbc2ebc5ca36560f43637730b2bfd7617b2a05a73404a292b91f69b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6f93c8399d3fc3448ac85ddf983ba51ceb74baed189e6b4632dc4e697a0046ac", "generation_seconds_batch": 23.524299837998115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 6.332377910614014, "genuineness": 4.394378662109375, "identity": 0.6338777542114258, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/09bdba9c1fbc2ebc5ca36560f43637730b2bfd7617b2a05a73404a292b91f69b.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "09f3f742a2fa37567623dbafff1e3bd147e62c1aa9d675faa5ff5ae9948a2482", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "124702bfd127cd7dc4a195e5a374ef5244232222e143064f9845743e3cdfaaee", "generation_seconds_batch": 17.002526212949306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.107357501983643, "target_z": 1.1328125080519105, "ce": 6.268836498260498, "genuineness": 4.7458367347717285, "identity": 0.6778165102005005, "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_S_MONO_reg/09f3f742a2fa37567623dbafff1e3bd147e62c1aa9d675faa5ff5ae9948a2482.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "09fb9566a85559f8c2c3a2e31c4760218de288bd2626fbbc6d12070da61267ca", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1151ed616a8964732e355075346c7ef2b4880e398845eb8c7bf99f2e9bc0f1f9", "generation_seconds_batch": 23.57065650902223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977931499481201, "target_z": 0.24511725127479017, "ce": 6.23706579208374, "genuineness": 4.457139015197754, "identity": 0.6387949585914612, "hypothesis": "Die Tür 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_S_MONO_reg/09fb9566a85559f8c2c3a2e31c4760218de288bd2626fbbc6d12070da61267ca.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0a4718da44633dcd44501d628ee8455571fdb43de52d0c5d94e0537ff8ad16e3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eb28dfa07b3f6a30eb8edeedfad18bbf5875d2bffaf8f1844627d866898a77a8", "generation_seconds_batch": 24.74641696899198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.893648624420166, "target_z": 0.9648436844867292, "ce": 6.062326908111572, "genuineness": 4.13078498840332, "identity": 0.6189854145050049, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0a4718da44633dcd44501d628ee8455571fdb43de52d0c5d94e0537ff8ad16e3.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0a5eae5a61512b1e83cd15da9f5bbb2443f0ea1ce1b059147fec2dae616f91b6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cb888dfd913f00db50d59f54e4cbdefd582cb3ce3d06a539dde58767a5d5e6ae", "generation_seconds_batch": 25.902563796029426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.308549880981445, "genuineness": 4.482243061065674, "identity": 0.7031381130218506, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0a5eae5a61512b1e83cd15da9f5bbb2443f0ea1ce1b059147fec2dae616f91b6.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "0a5ef9b4313677b5f6e28bb01b6bccf5347a470064b4ad63899df300a5a30cb8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d9a4657fcb5490d959ea9d9676d3a48a7fbf9890b505a3be2ad3455c8b181807", "generation_seconds_batch": 23.686179817072116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.933408498764038, "target_z": 0.9960937635418493, "ce": 5.911416053771973, "genuineness": 4.168440818786621, "identity": 0.5326875448226929, "hypothesis": "La puerta sigue que abierta. He dejado tu abrigo 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_S_MONO_reg/0a5ef9b4313677b5f6e28bb01b6bccf5347a470064b4ad63899df300a5a30cb8.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "0a61963c1a2808539f2a2898caba38e248f39082ddc4250358cd0a3a56f418e4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ad53ea63db6093145f895b8d71dab99929fafa52dadc465dba4c4d75373b959a", "generation_seconds_batch": 24.133278941968456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28482723236084, "target_z": 0.48632817239647247, "ce": 6.197352409362793, "genuineness": 4.030367851257324, "identity": 0.42280638217926025, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0a61963c1a2808539f2a2898caba38e248f39082ddc4250358cd0a3a56f418e4.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "0acd59ea6e86fe0c8873c3b345f511cee680898034ab0d16a7056d98bfc6a676", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bd766957c730a9c8f1657e5fc69fe088ff02151f15671f8de1b8ef226a28ac9", "generation_seconds_batch": 21.318227363051847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7890725135803223, "target_z": 0.09667970369531974, "ce": 6.205295085906982, "genuineness": 4.118232727050781, "identity": 0.5840999484062195, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0acd59ea6e86fe0c8873c3b345f511cee680898034ab0d16a7056d98bfc6a676.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "0b4387cac75e2b9a77a7c334f0a1105925a74b3bad42704f25ead92836876449", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5e3d09340adbde1eb48e714d658da03c0452d5550463d298066ef5b6bf9a5384", "generation_seconds_batch": 21.252716325921938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.23706579208374, "genuineness": 4.645420074462891, "identity": 0.7957607507705688, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0b4387cac75e2b9a77a7c334f0a1105925a74b3bad42704f25ead92836876449.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0b89696bb810d113d388e1e97bd88e1a2202d6dad3fab34945bb2d2359591ec4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "80e0f7051b7c69bbbc04745174eff96225d6a39fbb72ef4222248295194230c5", "generation_seconds_batch": 22.74904941197019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.504955291748047, "target_z": 1.4453125490434542, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.7468695640563965, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0b89696bb810d113d388e1e97bd88e1a2202d6dad3fab34945bb2d2359591ec4.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "0ba210399d92774a520bd65894ce47aaaef907b4f3bfa6601e22cb786f19cfcd", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c18acfe8745fb9d5da101079f4665e3817b07e17c11517e1b5684d5eb7eedfd4", "generation_seconds_batch": 17.058715209015645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 6.332377910614014, "genuineness": 4.7834930419921875, "identity": 0.47927942872047424, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0ba210399d92774a520bd65894ce47aaaef907b4f3bfa6601e22cb786f19cfcd.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "0c28fce14a084631707901699b7b4c18bb58ceab59def16fece1ad888a8148c8", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1f765351d084d6461f38321b0f6728d50f4a0d16995b7b4b9774fe06ecea812d", "generation_seconds_batch": 16.79256567498669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.141753673553467, "genuineness": 4.218649387359619, "identity": 0.689952552318573, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0c28fce14a084631707901699b7b4c18bb58ceab59def16fece1ad888a8148c8.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "0c2ef99c522f014990626819a5e68c81d36c998779dbff9bb93fcc19af4c9909", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "17289d203ff45935d8f28edb11143ef58a01ad6f07f42aac9a7ddb3e4a67d98d", "generation_seconds_batch": 21.278126031975262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0139636993408203, "target_z": 0.2734374890201222, "ce": 6.086154937744141, "genuineness": 4.494794845581055, "identity": 0.5584681034088135, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0c2ef99c522f014990626819a5e68c81d36c998779dbff9bb93fcc19af4c9909.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "0c4802052d0f4e3d86d725b49e59665c69a5ea701a1919af9c97991de999a5ab", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "431118d241e7a2c915ccbd36b8b07b990ecd6da7455bc31740b16517cb9c29ae", "generation_seconds_batch": 16.938079711981118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3022220134735107, "target_z": 0.49999995315252144, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.5009945034980774, "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_S_MONO_reg/0c4802052d0f4e3d86d725b49e59665c69a5ea701a1919af9c97991de999a5ab.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "0c9107c5408f933da1d4e23e296dc78c21510efd3ebfc8db9f50939b40ed128c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee604ae64ea5af0ff497b9d2c116dc24b148140b0d09412070593dd115d299b8", "generation_seconds_batch": 21.977638877928257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.205295085906982, "genuineness": 4.306513786315918, "identity": 0.7247928977012634, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0c9107c5408f933da1d4e23e296dc78c21510efd3ebfc8db9f50939b40ed128c.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "0cb65a37bc5372b1b82e2658ac41a5a25796f27a362b6bdf96824d9f3b0f2676", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2f18c3bf42d2d6e2bf7d9b6cd4234b675d56c4b28950d95dbfb6a760a838ef48", "generation_seconds_batch": 23.929362587979995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.514895439147949, "target_z": 1.4531252093496698, "ce": 6.046441555023193, "genuineness": 4.595211505889893, "identity": 0.7555071115493774, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0cb65a37bc5372b1b82e2658ac41a5a25796f27a362b6bdf96824d9f3b0f2676.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "37d0b1a8e0652bd083490257eabfea62230e9845f72ae13528e7b97e14df1453", "generation_seconds_batch": 16.563369904994033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.272402286529541, "target_z": 0.47656253440361707, "ce": 6.260893821716309, "genuineness": 4.2374773025512695, "identity": 0.7649294137954712, "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_S_MONO_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "0d4ce37ca465ef542c84b47d5498258b544718e50e76890d17f8be28e4eab7b2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7b0cfd0325592e30a25cfbed7bf97b4786951ecc2db36b925600868c27985b41", "generation_seconds_batch": 16.96149707690347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0288736820220947, "target_z": 0.2851562920895315, "ce": 6.300607204437256, "genuineness": 4.482243061065674, "identity": 0.709869384765625, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0d4ce37ca465ef542c84b47d5498258b544718e50e76890d17f8be28e4eab7b2.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "0d511ef4ab95202f3344973de3c08e75630a8cb88fbfc4d4593c535a8b354221", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ed667140ef7e96e0533fd6a49a4e294eec66a9c5cd027927021a36b523397954", "generation_seconds_batch": 23.709162737010047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 5.6334228515625, "genuineness": 4.645420074462891, "identity": 0.6279705166816711, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0d511ef4ab95202f3344973de3c08e75630a8cb88fbfc4d4593c535a8b354221.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "0d6643743e0a44cd5d38674837f865b6c4a79daf6d6db0a62d524f9e9ccb3fd5", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84a89f5e34266aca400c690d9285ced8b0b4749f4e89828af0602c79b97a30ed", "generation_seconds_batch": 16.95255221403204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4950151443481445, "target_z": 1.4374998887372383, "ce": 6.078212261199951, "genuineness": 4.375550270080566, "identity": 0.47863203287124634, "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_S_MONO_reg/0d6643743e0a44cd5d38674837f865b6c4a79daf6d6db0a62d524f9e9ccb3fd5.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "0daa3db87dcf371963f0e19efe43f3c5c53bbe256fcea1c2f297341145c1d2ff", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2a6faa3644847cb0cbba4de8b2d2c0cdd1a9563fe19f92ed02c7013cc4df81b5", "generation_seconds_batch": 20.62212675693445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 6.284721851348877, "genuineness": 4.3881025314331055, "identity": 0.5876796245574951, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0daa3db87dcf371963f0e19efe43f3c5c53bbe256fcea1c2f297341145c1d2ff.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "0dc0d353b0744fb6ab51c778eb9a186b179213de0b8d661fb02717c53a980bbc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1b184bdbcc2d03ca10be5bbb959e04f4bb74c5223b29328536507db0dd2b86d6", "generation_seconds_batch": 22.95910521398764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.32603645324707, "target_z": 1.3046876617701995, "ce": 6.229123115539551, "genuineness": 4.406930446624756, "identity": 0.6999083757400513, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0dc0d353b0744fb6ab51c778eb9a186b179213de0b8d661fb02717c53a980bbc.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "0dcd690d7b5fb9c08cf7d3f987e661b59734cc82a182ed6fd87d68f40e619470", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ff8b57d91aa25d667023e76e567e2c33f94ec1548187a7e7722479dfdd28d296", "generation_seconds_batch": 24.145741206943057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175487756729126, "target_z": 0.4003905955373279, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.5187947750091553, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0dcd690d7b5fb9c08cf7d3f987e661b59734cc82a182ed6fd87d68f40e619470.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0e20ffa28432db66e3d5bb1eb0c0adcc550b1c7f6ad90325a61e589e26d09bcc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0ebb4f854769302f6a1892ea7c253493bb8c2ef6bd369a46c062668b002a3856", "generation_seconds_batch": 24.34885942400433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.6915795207023621, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0e20ffa28432db66e3d5bb1eb0c0adcc550b1c7f6ad90325a61e589e26d09bcc.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "0eafa1523c5d2dc32158ef3f4127e533534ba23b9d8f0d6727d11871532066c3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1e7f7ea6fb9035b48c4412e8e1e570e636835b462ef3a2736ff7fe86a15e19b3", "generation_seconds_batch": 26.476698574027978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.704789638519287, "target_z": 0.8164061369072587, "ce": 6.316492557525635, "genuineness": 4.068024158477783, "identity": -0.01844797283411026, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0eafa1523c5d2dc32158ef3f4127e533534ba23b9d8f0d6727d11871532066c3.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "0f00e0d96fd2be9ede82a692f318217fde81f333e380c4832a4e0fac3b082875", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d2acd2f6f433fa6b4199a47a8bd4ac0f6b992967ce0be934404ed5d16f3743f1", "generation_seconds_batch": 18.557513216976076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197852611541748, "target_z": 0.41796870644648476, "ce": 6.149696350097656, "genuineness": 3.9425034523010254, "identity": 0.3256756663322449, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0f00e0d96fd2be9ede82a692f318217fde81f333e380c4832a4e0fac3b082875.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "0f046fd1afe0d48621f67714e111ac0912fe31250d96b4b815463c97eb313ef4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "21715a25377cc27c4942a19e7ea2cdcc6cf14fa1c36e8c5e22e0ad35650bcb3d", "generation_seconds_batch": 23.160940138972364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8787388801574707, "target_z": 0.9531250688072341, "ce": 6.252951145172119, "genuineness": 4.287685871124268, "identity": 0.6477577686309814, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0f046fd1afe0d48621f67714e111ac0912fe31250d96b4b815463c97eb313ef4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "0f91d0fb8be964ab855e05ddb67750e51b7740326a4c8267278189828e9e3f0b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4f1aa735d6d4358cb7daf4589e3f805c48e16d7c2026a9a30c93162dfc472570", "generation_seconds_batch": 22.40682736504823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.554655075073242, "target_z": 1.4843751010148756, "ce": 5.77639102935791, "genuineness": 4.645420074462891, "identity": 0.7089077830314636, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/0f91d0fb8be964ab855e05ddb67750e51b7740326a4c8267278189828e9e3f0b.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "101f7de86177687d2112c19b7a4c8f631d01f13e9608803d22f7739c2a035c36", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "16c538cf7bcce1c3f18efcd633512c4192b779450348234be43b6f587450f4fb", "generation_seconds_batch": 23.886859060032293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 5.974957466125488, "genuineness": 4.293961524963379, "identity": 0.4714395999908447, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/101f7de86177687d2112c19b7a4c8f631d01f13e9608803d22f7739c2a035c36.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "10e849e797054660ce1ccc13b6b4af3e87691198c196640c398395e6fba8bd74", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f748022f84075ea8aa9d38c3217fc5605e228b98a402124d1107bcc27361f754", "generation_seconds_batch": 17.333124459022656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.286276340484619, "target_z": 1.273437395325165, "ce": 6.221180438995361, "genuineness": 4.7834930419921875, "identity": 0.34516215324401855, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle comptait.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/10e849e797054660ce1ccc13b6b4af3e87691198c196640c398395e6fba8bd74.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "10eb4f417280a59798a13957d0e9f6ba12862df402d2082dd191c7540df91425", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4f6244daa0e7e6a1787e19ebb5ae14b49a9f4f22d03e3634a7905f31c332bf1c", "generation_seconds_batch": 17.185016855946742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.289797067642212, "target_z": 0.49023431515966603, "ce": 6.252951145172119, "genuineness": 4.381826400756836, "identity": 0.5327320098876953, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/10eb4f417280a59798a13957d0e9f6ba12862df402d2082dd191c7540df91425.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "10f0a4e499a10f443b36f2578886bacf7ce0fd2d402171e880ae3765f5029771", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4ac76150cc97faa71f11403acb5c21177d4edf5ad3d702b62bfd99bef4f0c22", "generation_seconds_batch": 20.599265680997632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.93837833404541, "target_z": 0.9999999063050429, "ce": 6.340320587158203, "genuineness": 4.457139015197754, "identity": 0.7439889311790466, "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_S_MONO_reg/10f0a4e499a10f443b36f2578886bacf7ce0fd2d402171e880ae3765f5029771.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "10f1d0015deb3e93cc201f9a38a20d872c61d88c41bf01bc16d95d9733786033", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1d6c78aeaa579c01a4d407c502f1378d4ac232d70b89004959e0cbb7e36dc2f7", "generation_seconds_batch": 23.811156190000474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.252951145172119, "genuineness": 4.8588056564331055, "identity": 0.5624474287033081, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/10f1d0015deb3e93cc201f9a38a20d872c61d88c41bf01bc16d95d9733786033.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "112396e97979a924ab346b320531dd51ae4e5ecca9a7cebf855a33cb756cc3b2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8160ef3bee030bdd3f59b4145012b4e111fcfa74792e1b6861e21af78ab13559", "generation_seconds_batch": 22.848785149049945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 6.475345611572266, "genuineness": 4.570107460021973, "identity": 0.3980242609977722, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/112396e97979a924ab346b320531dd51ae4e5ecca9a7cebf855a33cb756cc3b2.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "11e6bb7579a5f4fc970499c3e74bb186154c6d214ded5b5d81d393876481fbd0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1d5ec2cdf1fd6a381933912a776d093672afae7bbf33fce86ba9a1d01a72d3a4", "generation_seconds_batch": 26.170985282980837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.7138878107070923, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/11e6bb7579a5f4fc970499c3e74bb186154c6d214ded5b5d81d393876481fbd0.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "121aae1fc7867a9a0f8e756d4c28ebc4e02f25df8a19f0d40541febbdd32aac1", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "59d5a6828bc0a0876bca30e4553508da9a90cafa96c9072a355bb75f3e912eff", "generation_seconds_batch": 15.606256076018326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2823421955108643, "target_z": 0.4843750073199185, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.5461068749427795, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/121aae1fc7867a9a0f8e756d4c28ebc4e02f25df8a19f0d40541febbdd32aac1.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "1238329a27a11dba0e909f95ccdf0ed7ff1c3b4ab8651de99e51c9531f90fc35", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "108d665b0dd0c213a232188f724484e645e16b7e9130ff3a091c3c9fb3bf97a6", "generation_seconds_batch": 29.576803146046586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 5.808161735534668, "genuineness": 4.243753433227539, "identity": 0.6702108383178711, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1238329a27a11dba0e909f95ccdf0ed7ff1c3b4ab8651de99e51c9531f90fc35.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "3fa696025fd7e09e1cb357a6e2629da041f74927774cc259d9a74b52dd5fe92b", "generation_seconds_batch": 16.51340415002778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.2767791748046875, "genuineness": 4.268857479095459, "identity": 0.5958590507507324, "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_S_MONO_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "12b94ff4f088285e38a1aa51827cbd9c387c3fe18c8c59ed3b17aad8ddac65ef", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6329026979144ff30d84f72fd77d54d8c3ae8eafa60406afddba5e03e7b84e0e", "generation_seconds_batch": 17.465819238917902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7296395301818848, "target_z": 0.8359374128929695, "ce": 5.847874641418457, "genuineness": 4.218649387359619, "identity": 0.5642027854919434, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/12b94ff4f088285e38a1aa51827cbd9c387c3fe18c8c59ed3b17aad8ddac65ef.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "13890fb2021fb4cb6d7553ffdaf437a6eb45b3cab2bfe587588d69cc55d8f074", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "379b907bddb00b8c5777639d36c978bfd53595c067f8ed3ad8b5e63d8ea5f703", "generation_seconds_batch": 22.828428993932903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.814129114151001, "target_z": 0.9023437137664033, "ce": 5.796247482299805, "genuineness": 4.21237325668335, "identity": 0.5506734251976013, "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_S_MONO_reg/13890fb2021fb4cb6d7553ffdaf437a6eb45b3cab2bfe587588d69cc55d8f074.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "1394b656fbbf9738b7e62933eeddfd4d1ea14493a5dac5f36593212ade03cdab", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3a9bb84907f5ee6e936d2b9474915c7beec4c7c5ac9b7c75f88f09ace36e9edd", "generation_seconds_batch": 17.082865569042042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5656304359436035, "target_z": 0.7070311412992099, "ce": 6.356205940246582, "genuineness": 4.049196243286133, "identity": 0.532349705696106, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1394b656fbbf9738b7e62933eeddfd4d1ea14493a5dac5f36593212ade03cdab.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "13e690968abc2073bae5f130260a5866745bcc9459b47ef101e8ced556cb467e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "83112387b68cfe29078b9bfc2eb062b19b4d254b531c405fcc5fd01b31271dea", "generation_seconds_batch": 17.685997131979093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.3641486167907715, "genuineness": 4.344170093536377, "identity": 0.46059709787368774, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/13e690968abc2073bae5f130260a5866745bcc9459b47ef101e8ced556cb467e.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "1409db00940a0118f1741ed6c817a9bbb07385599cf5e2cbd3d7f7d8a7503e05", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9b64d5182c68664e8afaa386254ac499895370aaafb7b98750fcba9578689684", "generation_seconds_batch": 24.628098882036284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.001952171325684, "target_z": 1.8359373191980124, "ce": 5.919358730316162, "genuineness": 4.406930446624756, "identity": 0.6307798624038696, "hypothesis": "La porte 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_S_MONO_reg/1409db00940a0118f1741ed6c817a9bbb07385599cf5e2cbd3d7f7d8a7503e05.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "14d62c05ddcfa38e4c6db3c600f33a82920ad1c77c350e8f702a2dd028364b62", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "40b46e355d633c29d71aa2e7818b278a37278bd81ae2a10bee9e36f14da1a2c0", "generation_seconds_batch": 17.024466399103403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.24500846862793, "genuineness": 4.595211505889893, "identity": 0.7241873741149902, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/14d62c05ddcfa38e4c6db3c600f33a82920ad1c77c350e8f702a2dd028364b62.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "14e1b51f20e719db6498d546e7b75241738584f018018137f9b735061ee6a4df", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ff3a065420e46804b9b6c5ecdfc88c7852a06546cb7bfb96abde3b629178502c", "generation_seconds_batch": 18.32607819302939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4364113807678223, "target_z": 0.6054688059973767, "ce": 6.157639026641846, "genuineness": 4.519899368286133, "identity": 0.6486213207244873, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/14e1b51f20e719db6498d546e7b75241738584f018018137f9b735061ee6a4df.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "14e7a37b6c1ec08a839d36f4e9db85c6221bb77fe52341ac8376e4b347233ff8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4cb676e94abe0e2d3a1e64179e630852ac2626593df117876c7dd95c9e6a4dd5", "generation_seconds_batch": 16.78835791500751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.0146708488464355, "genuineness": 4.287685871124268, "identity": 0.5872657299041748, "hypothesis": "De deur 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_S_MONO_reg/14e7a37b6c1ec08a839d36f4e9db85c6221bb77fe52341ac8376e4b347233ff8.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "14f3d84b140e8f184ae8a345605f39ae697c3e53cac7d47d64dc866d54b9ff5e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0d2782e3650337c08f4376f989beb55b1d90b19cdf94aa76c3dffcfc2190d1f3", "generation_seconds_batch": 22.991529899067245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.923468589782715, "target_z": 0.9882812906255478, "ce": 6.300607204437256, "genuineness": 4.381826400756836, "identity": 0.6548417806625366, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/14f3d84b140e8f184ae8a345605f39ae697c3e53cac7d47d64dc866d54b9ff5e.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "cb244658da5c5356d6125e3532ee4b94cd67d5bd85f239b55942c10672fc1439", "generation_seconds_batch": 19.124837178969756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5159308910369873, "target_z": 0.6679687767177026, "ce": 6.1894097328186035, "genuineness": 4.595211505889893, "identity": 0.6560689210891724, "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_S_MONO_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "156ad4d04b9437cf4d41fe7c9a38062710601c690e1827b8eb3246e0af2d3bc7", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "93c756534df5fa31987078b5684d4827da8723e558bf6a7db9ba68858ad20d94", "generation_seconds_batch": 16.862392148002982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.819099187850952, "target_z": 0.9062500439195111, "ce": 6.229123115539551, "genuineness": 4.2751336097717285, "identity": 0.7596763968467712, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/156ad4d04b9437cf4d41fe7c9a38062710601c690e1827b8eb3246e0af2d3bc7.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "15abd01856462d4a6861abe31c3a8910dcbbbd2e24dda0add8f38d8d139dd5a0", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e335ff7f0c0f5d91dc6f16438ad7eb418fec87a97ec36b2b0bed8433ab6d8e74", "generation_seconds_batch": 17.044571025995538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.118333101272583, "target_z": 0.3554687357261589, "ce": 6.372091293334961, "genuineness": 4.607763767242432, "identity": 0.2659923732280731, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las ze twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.120000000000001, "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_S_MONO_reg/15abd01856462d4a6861abe31c3a8910dcbbbd2e24dda0add8f38d8d139dd5a0.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "15c5258894827b94ce8367011f33bf55d253635b4e10696f7a32827fcc250c49", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "42bf040095ce636b104c8a9f37ae9041ad5bcafc902d69f2e85edf758c132a99", "generation_seconds_batch": 21.535584743018262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8290390968322754, "target_z": 0.9140625168358126, "ce": 6.41180419921875, "genuineness": 4.356721878051758, "identity": 0.5877980589866638, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/15c5258894827b94ce8367011f33bf55d253635b4e10696f7a32827fcc250c49.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "161ccdea91b5620d3be4f8d18cbaf96abe9466771bff2a630e5990dc5ad7e01e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "624f0c9b468b7368f0602ec240f35abfdaa6f68e0b5910a51346c958bcdb5af9", "generation_seconds_batch": 22.944575543981045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.5436480045318604, "hypothesis": "Die Tür 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_S_MONO_reg/161ccdea91b5620d3be4f8d18cbaf96abe9466771bff2a630e5990dc5ad7e01e.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "162d3d9d3633516b385a3ca78ac60012d06b75f68fa03d9fbe9aad80b9d07df5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c2690c65975be62239b55bd91d2496b951a942cb0cf669ad2f6c882b735ed857", "generation_seconds_batch": 23.51376200595405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.634174346923828, "target_z": 1.5468748843452873, "ce": 6.133810997009277, "genuineness": 4.833701133728027, "identity": 0.6340471506118774, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/162d3d9d3633516b385a3ca78ac60012d06b75f68fa03d9fbe9aad80b9d07df5.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "16570d5264e1b5620e1afcee254f22754a750dc3ca64843d8157f4956aa31ab4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7df80abc7cb59c4b2862bc6d59143f237c3fd5971d28f2e69c13f94c258ee00b", "generation_seconds_batch": 17.25613145006355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3767716884613037, "target_z": 0.5585937811096537, "ce": 6.292664527893066, "genuineness": 4.168440818786621, "identity": 0.6577988862991333, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/16570d5264e1b5620e1afcee254f22754a750dc3ca64843d8157f4956aa31ab4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "165f0fffa06b0a1ff7e8eb354df6f0ac17a03789644c38c36811068be0e88bb6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e7d230fc7537fd6f77ac5d228e731684ff8c58f8c91c6582eed969f79d63e5ab", "generation_seconds_batch": 22.009716066066176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.848918914794922, "target_z": 0.9296874626684155, "ce": 6.300607204437256, "genuineness": 4.469690799713135, "identity": 0.5804753303527832, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/165f0fffa06b0a1ff7e8eb354df6f0ac17a03789644c38c36811068be0e88bb6.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "168eebc7010e7839440d7a4906d3921794d9e6c46036eb1afad4ed66a12f686b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3cde0f5c0e8efb8c1093e19e275653d9c4d33cf2050c739e7069094eace65185", "generation_seconds_batch": 23.67650514608249, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.317131996154785, "target_z": 0.5117187562219307, "ce": 6.078212261199951, "genuineness": 4.632867813110352, "identity": 0.733723521232605, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/168eebc7010e7839440d7a4906d3921794d9e6c46036eb1afad4ed66a12f686b.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "16b4da04feca8e8b9614f42a448b27b127b7eb7f549314c5c8fb74e3b74aa63b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e274c9cb26a9772fecda0b16461dfe247a5bf3211e923de5771b7caa64ba5e5d", "generation_seconds_batch": 22.215440520085394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9135284423828125, "target_z": 0.9804686303193321, "ce": 5.879645347595215, "genuineness": 4.68307638168335, "identity": 0.7703765630722046, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/16b4da04feca8e8b9614f42a448b27b127b7eb7f549314c5c8fb74e3b74aa63b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "1728964ec4d7378b7ea0173f9f343d81b80e63f0ace7af32e329a9c74518dcac", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "045cd9038712e2ed1268246c7078ea7a321d646210c375cacd8955627a590e58", "generation_seconds_batch": 23.98368084896356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4463512897491455, "target_z": 0.6132812789136782, "ce": 6.395919322967529, "genuineness": 4.469690799713135, "identity": 0.5364679098129272, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1728964ec4d7378b7ea0173f9f343d81b80e63f0ace7af32e329a9c74518dcac.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "17394e9d9a7c12765ab1f3a1f6ff4e37f8d0cbf6b37dc2b555e5bfb92d990e18", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6d9228767f8e7ef91729819562f5ed0f44f1d951abd8bb84449a35c215be020c", "generation_seconds_batch": 16.056258635013364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.229123115539551, "genuineness": 4.3127899169921875, "identity": 0.6981991529464722, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/17394e9d9a7c12765ab1f3a1f6ff4e37f8d0cbf6b37dc2b555e5bfb92d990e18.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "17fa28f540f243d34f059d2084a09eeb08e3f5a890d41e67c58318839c4ff269", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2200e1f74d1100821086b9de9f643a55c6a4201f75e62bc7602d1b8aa592eb7c", "generation_seconds_batch": 27.39263319107704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.141753673553467, "genuineness": 4.519899368286133, "identity": 0.47934964299201965, "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_S_MONO_reg/17fa28f540f243d34f059d2084a09eeb08e3f5a890d41e67c58318839c4ff269.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "18434d3f5c3aea32ea1d812b3dafe63e12b1807ac581e22456218655765bd4e1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d6964b5afaff53652bee7b3f7c7ecc64d1f134fcd8dd4c3ab21752194946e991", "generation_seconds_batch": 21.535584743018262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.5908685922622681, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/18434d3f5c3aea32ea1d812b3dafe63e12b1807ac581e22456218655765bd4e1.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "18754826c1f23159f3c371c7388828616f35bc84686d1857bae90372c51848e8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "917f87efd438346568fdd2eeb899146041fcfcb6ff2422c7aef8a2658710f2cb", "generation_seconds_batch": 17.39513406693004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4237794876098633, "target_z": -0.1904297534707657, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.4945789575576782, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/18754826c1f23159f3c371c7388828616f35bc84686d1857bae90372c51848e8.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "187d6e6189dc2bb7a0a446f0f7cfdfcdf40b6df8985b9d84f80dee575b5b3175", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f7ced74309cbbc21e7c5bbf2e951960e2633c54a8f2aeb47d22ebb7e005b2f09", "generation_seconds_batch": 24.215460938983597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.3005162477493286, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/187d6e6189dc2bb7a0a446f0f7cfdfcdf40b6df8985b9d84f80dee575b5b3175.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "18847f5f94e13a4940764077c28d88f58f742571eae8592f7e65e5bc60bdbbd2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "92baf90120d662750dbbe022bc700c3ed947173de5a933765900cb58c0abe3ce", "generation_seconds_batch": 22.64624836598523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.054384231567383, "genuineness": 4.017816066741943, "identity": 0.5145992636680603, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/18847f5f94e13a4940764077c28d88f58f742571eae8592f7e65e5bc60bdbbd2.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "18b0f0379c57565705634f22b2c39edbf44733547ff01d651f73c25da25e090b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5202ba6c6defcc0358e55d04e1fb0b553f7943b38ca41493b207c4d3fb889539", "generation_seconds_batch": 16.78835791500751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2823421955108643, "target_z": 0.4843750073199185, "ce": 6.062326908111572, "genuineness": 4.507347106933594, "identity": 0.6552053689956665, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/18b0f0379c57565705634f22b2c39edbf44733547ff01d651f73c25da25e090b.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "18e955a2a5cb85715755257495c70c0908b8596d6eef5da2f68ec2ebee2a1ca9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "74e61cef8bd648a1f1f063937b61aa4cf0dcce037554866728ae4006216b4bb2", "generation_seconds_batch": 25.27702296501957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8340091705322266, "target_z": 0.9179688469889205, "ce": 6.205295085906982, "genuineness": 4.394378662109375, "identity": 0.530685544013977, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/18e955a2a5cb85715755257495c70c0908b8596d6eef5da2f68ec2ebee2a1ca9.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "1a321ea7b45a196e28577f2c7a4067935ffda3d05217dde884544fe0927f7801", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "15e4efa5c9ee29536ebd6b24f87654a83723f157b67f0d51b9c9ee86a7131c48", "generation_seconds_batch": 22.28933443292044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.2662995457649231, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1a321ea7b45a196e28577f2c7a4067935ffda3d05217dde884544fe0927f7801.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "1a91e5dcdf31489c0470160d884fab326f0cb57c126ec3c25ec57e377329f72c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dc378bbd9bad6abacf96c3bc617757170a2919dc18a281c2fc05be8996785f26", "generation_seconds_batch": 24.70726302091498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.554655075073242, "target_z": 1.4843751010148756, "ce": 6.086154937744141, "genuineness": 4.871357440948486, "identity": 0.7158521413803101, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1a91e5dcdf31489c0470160d884fab326f0cb57c126ec3c25ec57e377329f72c.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "1b18b269b27c149722a952faa33d983eb44e828c1b0165af84b38450cd247ce2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "62579904f4103aae7688ca2dc36d2fcdb338839bf5da130c2c8826090a99e517", "generation_seconds_batch": 16.90658159099985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087477684020996, "target_z": 1.1171875622193075, "ce": 6.340320587158203, "genuineness": 4.218649387359619, "identity": 0.32086795568466187, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1b18b269b27c149722a952faa33d983eb44e828c1b0165af84b38450cd247ce2.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "1b4706ab58e5fb520255e31303c788af398761f99d365400ee030879ce86cecc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cbcc768ae7c1a15357506a30a046c19b3da6f891a6e8f7dbae58127eda6077b6", "generation_seconds_batch": 21.682594570913352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 6.3641486167907715, "genuineness": 4.545003414154053, "identity": 0.6017633676528931, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1b4706ab58e5fb520255e31303c788af398761f99d365400ee030879ce86cecc.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "1b7f7ca0f61396af113a6fed3ffb74029f1a085991f3198995d44d276c673045", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "89d000d18154e26027a2aae363fbab17ae88096d8f23204e2189bc65c8561b6c", "generation_seconds_batch": 17.850500177009962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.7741883397102356, "hypothesis": "La porte 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_S_MONO_reg/1b7f7ca0f61396af113a6fed3ffb74029f1a085991f3198995d44d276c673045.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "1ba06dc958cc6fe1ccd57ca6fd3825c644e0eb1c140de119db49e5739048bc60", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "232300e4ad2f50a66259874028d517bf07e141157a6ff695aac47650a61f1fa0", "generation_seconds_batch": 17.95903439295944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 5.629451274871826, "genuineness": 4.306513786315918, "identity": 0.7086337208747864, "hypothesis": "J'ai trouvé la petite boite. Elle était près de la vieille fenêtre.", "wer": 0.42105263157894735, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1ba06dc958cc6fe1ccd57ca6fd3825c644e0eb1c140de119db49e5739048bc60.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "1bf6441b7b5b0c7d4b69c6c9e949db2ad302f08770f4aa58515908336f72165c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bcae61609f0f2f5b71d8d84851d805dcf1e3ad25c813ed20af1ec39316088e75", "generation_seconds_batch": 23.988912081927992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6103603839874268, "target_z": 0.7421875505074378, "ce": 5.919358730316162, "genuineness": 3.9927117824554443, "identity": 0.6754436492919922, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1bf6441b7b5b0c7d4b69c6c9e949db2ad302f08770f4aa58515908336f72165c.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "1c21e64614595c34b31d16f2239b87136e237813ba9f050a4762b17cab5dfed5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "585198732b3c78fd52977de19d82b8981fbe414e585e2f4519d2cf40a330c7d2", "generation_seconds_batch": 20.313041946967132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.078212261199951, "genuineness": 4.65797233581543, "identity": 0.7080570459365845, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1c21e64614595c34b31d16f2239b87136e237813ba9f050a4762b17cab5dfed5.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "1c25d05333f3215f88e4e41ceb3a709e3ffc2a1f24d7677d03b237d1e411921e", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3055d8af955d0bb95417e1fff2323725f00261268d6858a7e6ca645e108b4fe4", "generation_seconds_batch": 15.670681391027756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9667489528656006, "target_z": 0.2363281021252546, "ce": 6.348263263702393, "genuineness": 4.444586753845215, "identity": 0.43536731600761414, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1c25d05333f3215f88e4e41ceb3a709e3ffc2a1f24d7677d03b237d1e411921e.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "1ca2134d262f310d2f5f9243b87fff56ea7980eeda348ea66a516912c1259387", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "46a60a04fa2e89edf9cf54f3b7909403bfcd63c701f013008d15dd44b1688007", "generation_seconds_batch": 24.305325884954073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.292664527893066, "genuineness": 4.582659721374512, "identity": 0.6367758512496948, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1ca2134d262f310d2f5f9243b87fff56ea7980eeda348ea66a516912c1259387.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "64725a3b64e747649b7b403931f5bf0b6541d9c20353ff0179e39b72a6fe4ce1", "generation_seconds_batch": 16.738204400986433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3767716884613037, "target_z": 0.5585937811096537, "ce": 6.300607204437256, "genuineness": 4.519899368286133, "identity": 0.7467113137245178, "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_S_MONO_reg/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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "88e1fdbaf3d1ece2779e00be2d3d9834167a45a665c30dd60254602e61c251aa", "generation_seconds_batch": 14.915647619985975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687188148498535, "target_z": 0.01660162206583701, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.5868452191352844, "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_S_MONO_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "1dd58c8e7f798a47de518c1e51411d205eb32883320270d77736cd7bb8e3d2ec", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "39cecc75760cb443b7bbb424a39c445f70913e4cbf24c8096d611e160b315797", "generation_seconds_batch": 15.694387238938361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.23761248588562, "target_z": 0.4492187855016049, "ce": 6.149696350097656, "genuineness": 3.8797430992126465, "identity": 0.16232597827911377, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1dd58c8e7f798a47de518c1e51411d205eb32883320270d77736cd7bb8e3d2ec.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "1e2085a6df1dfe6f1d42c40f444574053c38d60b4a922fdad470477ab3bd229f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "762b25cdf9975d3bea838de626157660536c5839f17e3b92a3fe94be06289401", "generation_seconds_batch": 16.342151918099262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.761737585067749, "target_z": 0.07519526263305504, "ce": 6.308549880981445, "genuineness": 4.319066047668457, "identity": 0.6305067539215088, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1e2085a6df1dfe6f1d42c40f444574053c38d60b4a922fdad470477ab3bd229f.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "1e5281e1088789f3b42fb1e824620408a10bdfdad8965954fa64f2a5823a9e4a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "09ed39622efa163184930f3434107cf9d80b7897e605d8f8786d17ee5c118b0f", "generation_seconds_batch": 19.951789537095465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.316492557525635, "genuineness": 4.519899368286133, "identity": 0.6546039581298828, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1e5281e1088789f3b42fb1e824620408a10bdfdad8965954fa64f2a5823a9e4a.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "1e8f82efe7c2c72220ef6775fe22de9305b76f9174913ef432635c76b88217b4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2c354faaa17d7a37d4b8212e746144f6df919ed62845a2706aa5a8e422c22b7d", "generation_seconds_batch": 22.702436299063265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.186877250671387, "target_z": 1.1953126661621505, "ce": 6.2767791748046875, "genuineness": 4.256305694580078, "identity": 0.7819383144378662, "hypothesis": "De deur 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_S_MONO_reg/1e8f82efe7c2c72220ef6775fe22de9305b76f9174913ef432635c76b88217b4.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "1ea13e36c1b0757df560316661162da507d0c4d287f2de091e064e2dc3a17995", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e8b921dd2f09bd3daae012abbef5b75c2cd502e8d397a68f823215c1c227feae", "generation_seconds_batch": 25.741844723932445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.100938320159912, "target_z": 0.3417969549701099, "ce": 6.284721851348877, "genuineness": 4.2751336097717285, "identity": 0.41096949577331543, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1ea13e36c1b0757df560316661162da507d0c4d287f2de091e064e2dc3a17995.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "b360e52841482c71e0b4d2c20567a2cd3c9de1c783e04b9e64b3c3360b410365", "generation_seconds_batch": 16.738204400986433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665030002593994, "target_z": 0.785156245242053, "ce": 6.284721851348877, "genuineness": 4.268857479095459, "identity": 0.7332629561424255, "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_S_MONO_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "1ee6c7489a6d947a7323c7cf35f5b720ab3ab4b562efcc8a717d5ef84995cdc3", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54aa7082b7fcd1963f52293363c02bd5e2f890b32489705aa5206e6e58a07200", "generation_seconds_batch": 27.39263319107704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.38003396987915, "genuineness": 4.570107460021973, "identity": 0.4777499735355377, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch toon.", "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_S_MONO_reg/1ee6c7489a6d947a7323c7cf35f5b720ab3ab4b562efcc8a717d5ef84995cdc3.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "1f6e867b2704b65b5e482afbec0cb49cfea04338c3a2937026f693df1565a552", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1eccb35edf6d8bb95a9c7c2637b7cf2cf49d2112264b2f7f15fa14f43b546605", "generation_seconds_batch": 21.706229066010565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 6.229123115539551, "genuineness": 4.143336772918701, "identity": 0.18063586950302124, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1f6e867b2704b65b5e482afbec0cb49cfea04338c3a2937026f693df1565a552.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "1fbacaf66c97e3378aa6854e4cdfa1da1043e650f701460cd5ddf37cba1efdfc", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c0478aa9affce4e56f08be6aa9b6afc4ca203f3741e8f80169d1bb81198f2c4a", "generation_seconds_batch": 18.41823343699798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2525224685668945, "target_z": 0.4609375885710142, "ce": 6.348263263702393, "genuineness": 4.3504462242126465, "identity": 0.4985003173351288, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1fbacaf66c97e3378aa6854e4cdfa1da1043e650f701460cd5ddf37cba1efdfc.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "1fcfdf78effd8f45682538ffa1c904013b499089bf12d3069bbaf54c8dc96423", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7e237668ddc99facc1edc6bb70072932884b3bf04a659106d32dab37be66912c", "generation_seconds_batch": 15.550295089022256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8387722969055176, "target_z": 0.13574225566674128, "ce": 6.054384231567383, "genuineness": 4.030367851257324, "identity": 0.7205405235290527, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/1fcfdf78effd8f45682538ffa1c904013b499089bf12d3069bbaf54c8dc96423.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "20865d927c09ee7158fade9ab74ca2e5687a3392df3f817a8224de5ed237067f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2b8f356e9c98693ef6a41bdd8b828142710d70d57c0aef4583bb07929e86739b", "generation_seconds_batch": 22.633633731980808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.42647123336792, "target_z": 0.597656145691161, "ce": 5.855817794799805, "genuineness": 4.381826400756836, "identity": 0.6730883121490479, "hypothesis": "Die Tür 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_S_MONO_reg/20865d927c09ee7158fade9ab74ca2e5687a3392df3f817a8224de5ed237067f.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "20af1f704d4030029dadd6737b93ef71052a5a2c04a0b71f8cf592cfe4dacee6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6f11d90dc8df804ec0d925b2f5f6936d9ae8fdf466e601351572da5f987d523c", "generation_seconds_batch": 23.256876711035147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.8039021492004395, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/20af1f704d4030029dadd6737b93ef71052a5a2c04a0b71f8cf592cfe4dacee6.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "211264c1218d2910086755af312e1fbb2fef90f367c2465554e2854576828fdc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a465522c6412679576b687f915df319d3671f0f90ccab105f72f387b02c1c15e", "generation_seconds_batch": 27.424577189958654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.431441307067871, "target_z": 0.6015624758442688, "ce": 6.308549880981445, "genuineness": 4.281409740447998, "identity": 0.4518238306045532, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/211264c1218d2910086755af312e1fbb2fef90f367c2465554e2854576828fdc.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "215553a6209f92b9754412eb4c9f52b43b94309f59dcfa7ecae3f0e5a712db9a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bf4a8b270b77138ca8bb6beea01518a899e7fa8f67537567366a961c34d30151", "generation_seconds_batch": 24.74641696899198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.09409761428833, "genuineness": 4.306513786315918, "identity": 0.6871873140335083, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/215553a6209f92b9754412eb4c9f52b43b94309f59dcfa7ecae3f0e5a712db9a.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "21a23f5976f1d226c3c04a0040ee7e5fe068670ec1d7d540825f504d7c560434", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "792c377189a2f229fed521460ad595dbc7dbeebd6d03846e1bc25044126bdac2", "generation_seconds_batch": 23.60147459502332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.893648624420166, "target_z": 0.9648436844867292, "ce": 5.855817794799805, "genuineness": 3.973883628845215, "identity": 0.6423067450523376, "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_S_MONO_reg/21a23f5976f1d226c3c04a0040ee7e5fe068670ec1d7d540825f504d7c560434.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "21b04ccf9a2e4a775bb824eac4d2ecb15976806f4cc9cfecf1a764f1ee5d9c70", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "60b8ae33b26abb1b01456cdf29533656cd52386a62673fe8492790c71e7496b3", "generation_seconds_batch": 17.133620452019386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8439488410949707, "target_z": 0.9257811325153077, "ce": 6.078212261199951, "genuineness": 4.519899368286133, "identity": 0.4906693994998932, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/21b04ccf9a2e4a775bb824eac4d2ecb15976806f4cc9cfecf1a764f1ee5d9c70.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "21fbc5fe0619936bf6d22a5457431749915de116b4a39433c49c9725c227c1a4", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a068cf984208d20c098cf24d6eb21a1269664f255262072752e5a75a56fd51e", "generation_seconds_batch": 16.890796623076312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9035887718200684, "target_z": 0.9726563447929449, "ce": 6.23706579208374, "genuineness": 4.758388519287109, "identity": 0.5640360116958618, "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_S_MONO_reg/21fbc5fe0619936bf6d22a5457431749915de116b4a39433c49c9725c227c1a4.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "2332c351a41eee76fbb9ccf3ec1c20c9476dd8aac50f119a6055675f1271b4d0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c31d7be006a7e70ad8bae3b433dc6086f16ecb683ed25e5e24b6fcc5075db035", "generation_seconds_batch": 24.442791369045153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.348263263702393, "genuineness": 4.469690799713135, "identity": 0.7088932394981384, "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_S_MONO_reg/2332c351a41eee76fbb9ccf3ec1c20c9476dd8aac50f119a6055675f1271b4d0.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "247cc32f9c1330a384fe2524a28a143a6783827e8f2940192b251ea5225b803a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "63dadca213c1f0dd5f6a92e575596125ac5e3f8dd85d257d558006bf883f58b9", "generation_seconds_batch": 21.706229066010565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.317131996154785, "target_z": 0.5117187562219307, "ce": 6.157639026641846, "genuineness": 3.9048471450805664, "identity": 0.3262075185775757, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/247cc32f9c1330a384fe2524a28a143a6783827e8f2940192b251ea5225b803a.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "24f95cc63db6487c5d7d439d8278d0a471287915489bc6d342bd4dd04070d6bc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4d775d070eada4d141d20dcb99ed481486b9f387007b3d0d05a831e6e33dd1aa", "generation_seconds_batch": 24.05934224196244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.1894097328186035, "genuineness": 4.557555198669434, "identity": 0.615105390548706, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/24f95cc63db6487c5d7d439d8278d0a471287915489bc6d342bd4dd04070d6bc.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "250447c0ad917a514f47c2410e9b3b2d73546f3ce8ef3b09a7b5567cb921cba3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "200b97be86450dd14004b52f44d65be1efde925a0b08664d6f48f9cce3c28633", "generation_seconds_batch": 22.264705550973304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.077537536621094, "target_z": 1.1093749019130918, "ce": 6.117925643920898, "genuineness": 4.607763767242432, "identity": 0.6675916910171509, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/250447c0ad917a514f47c2410e9b3b2d73546f3ce8ef3b09a7b5567cb921cba3.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "252f4df07919f13a0c073f04b1274fc32aef8411d3a6fc4ffd78e2b3837bb7f0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "88f7ca81c73680f388b6d6d3691be7f418dba3c0fd9f101aa553b3270f9c7fff", "generation_seconds_batch": 17.02790195296984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9880781173706055, "target_z": 1.0390624582764645, "ce": 6.213237762451172, "genuineness": 4.432034492492676, "identity": 0.6174355745315552, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/252f4df07919f13a0c073f04b1274fc32aef8411d3a6fc4ffd78e2b3837bb7f0.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "266cf64f14d427b1c07c33eb31b1bb84047309ce0106952d8b6b363f75d0990b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 130, "seconds_generated": 10.4, "mp3_sha256": "ceddb4ceb53cb907c4a5074d0bda82457129b6355a568d38281faf902a2f12e3", "generation_seconds_batch": 25.13087231805548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.22663688659668, "target_z": 1.2265625578273565, "ce": 5.506340026855469, "genuineness": 4.042920112609863, "identity": 0.4256429672241211, "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_S_MONO_reg/266cf64f14d427b1c07c33eb31b1bb84047309ce0106952d8b6b363f75d0990b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "26a754a284bdf2480634725e4a842203f30305c2bbd0117b236f0ce5eae9c0f3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9d2ea2f0b48e264daf6715168b9d6ae5fefedf0ad1cad14c343e7d831934ef64", "generation_seconds_batch": 23.962809504009783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 6.229123115539551, "genuineness": 4.557555198669434, "identity": 0.6370836496353149, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/26a754a284bdf2480634725e4a842203f30305c2bbd0117b236f0ce5eae9c0f3.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "26ba62c0dd731386dbf20ffc68ec44101b3bd7a2fa0d0b79f9a916ff7669b0af", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c5e34e64bdc04e26bd1b82953dafffb9d4842f3be04185b2c7fc7d1b3659a85f", "generation_seconds_batch": 22.943371178931557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.22021746635437, "target_z": 0.4355468173556416, "ce": 6.0305562019348145, "genuineness": 4.482243061065674, "identity": 0.7386929392814636, "hypothesis": "De deur 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_S_MONO_reg/26ba62c0dd731386dbf20ffc68ec44101b3bd7a2fa0d0b79f9a916ff7669b0af.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "004a2465e709dc1d4f88307769535deaa00197d32ec7c4fb089797be96bd8491", "generation_seconds_batch": 17.752049432019703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0686333179473877, "target_z": 0.31640618375473734, "ce": 6.41180419921875, "genuineness": 4.356721878051758, "identity": 0.5072293281555176, "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_S_MONO_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "27080b9b188c42b18940df8a1c79ca6adf2a3d4e1f0c732b687068a25ab26dea", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5753cb5601f830e4d8c0f387673003a6fa6eed7e7871e8f52bfbc7e3bf364ea", "generation_seconds_batch": 20.55550511495676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.77933931350708, "target_z": 0.874999964864391, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.5748735666275024, "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_S_MONO_reg/27080b9b188c42b18940df8a1c79ca6adf2a3d4e1f0c732b687068a25ab26dea.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "272f09bef54859f0c6102d0151fa8a0013650f776087bb21c336934e2cf6bd1f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cac815d3b1e38335ac1b3265eb61faed5ea03d25cf7d6e41cf3a9d12f67df590", "generation_seconds_batch": 23.08076905296184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.483288764953613, "genuineness": 4.268857479095459, "identity": 0.5154311656951904, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/272f09bef54859f0c6102d0151fa8a0013650f776087bb21c336934e2cf6bd1f.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "273e61cf30805367bb3c8239af07916fb024069e5e221c8ea9bf9517d2d4690f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "346925fbbb946f30c209664197d6342a6b9f86fa0d9450f3a289d2f40e2a2b9e", "generation_seconds_batch": 16.526820008992217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.968198299407959, "target_z": 1.0234375124438615, "ce": 6.260893821716309, "genuineness": 4.425758361816406, "identity": 0.5048348307609558, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/273e61cf30805367bb3c8239af07916fb024069e5e221c8ea9bf9517d2d4690f.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "2747729cf035ad2ee031d0dd20c0371de1ade7b300d071aef383217310ba1c68", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "7e58e90f23722dbeb11e839e996e191324169bca36d6b2674d0e6110b3b5aba3", "generation_seconds_batch": 21.78256533597596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.574534893035889, "target_z": 1.5000000468474786, "ce": 5.669164657592773, "genuineness": 4.319066047668457, "identity": 0.6608079075813293, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2747729cf035ad2ee031d0dd20c0371de1ade7b300d071aef383217310ba1c68.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "2763bc230c2f72c35e6ca0ef8746bef490326610d0099257aede413fdb350608", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e415790c8a238fb97378f570003d04a74a54d16f7ad5cb41053e232644176028", "generation_seconds_batch": 16.832905864110216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8290390968322754, "target_z": 0.9140625168358126, "ce": 5.967014789581299, "genuineness": 4.419482707977295, "identity": 0.3003188371658325, "hypothesis": "La puerta sigue abierta. He dijiado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2763bc230c2f72c35e6ca0ef8746bef490326610d0099257aede413fdb350608.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "27f0b1925c5ce63ec6801c68e964f7dea266ea549fc77ed443dd421e446143ff", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "990980e4df4920190994b2f8cccfec7f349d721ff64514e775af5d18840c30d2", "generation_seconds_batch": 16.946800643927418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1382129192352295, "target_z": 0.37109368155876177, "ce": 6.403861999511719, "genuineness": 4.419482707977295, "identity": 0.3991820812225342, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/27f0b1925c5ce63ec6801c68e964f7dea266ea549fc77ed443dd421e446143ff.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "281fe72e7bf1503accc3d95b81a57f048e1ec17b8f7a0a6873f500461a1104cb", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "90c4ddb806e495a9fff921c75ff09419ac1f02d2833d1a46a58b85763c35dfb2", "generation_seconds_batch": 18.687576682073995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1332430839538574, "target_z": 0.3671875387955682, "ce": 6.054384231567383, "genuineness": 4.243753433227539, "identity": 0.6016420722007751, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/281fe72e7bf1503accc3d95b81a57f048e1ec17b8f7a0a6873f500461a1104cb.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "28956cf96d91f97ac1a7cc4d07964b864f660d2322c96ac6fd9c7d323bbc86c6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "11b42b1ffef27935010c5cbde8809b3159172e89ed40d77d2bec0e8b6e7728cd", "generation_seconds_batch": 25.31284831801895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.202822685241699, "target_z": 0.42187503659959263, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.4462541937828064, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/28956cf96d91f97ac1a7cc4d07964b864f660d2322c96ac6fd9c7d323bbc86c6.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "28f2fff4f7d9f39de1cb0a97c06307dbb7fdf574828c59bb3fb4f80d7176e740", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e5be4b63871a59a7e9f1112346300cb362a3326dbc6fedae3dd461c7dc35f3b5", "generation_seconds_batch": 23.770071857026778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1456680297851562, "target_z": 0.37695317678842355, "ce": 6.332377910614014, "genuineness": 4.557555198669434, "identity": 0.5260072946548462, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/28f2fff4f7d9f39de1cb0a97c06307dbb7fdf574828c59bb3fb4f80d7176e740.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "28f464be66d74182727ea12b7f64e1c06295dee817d94d7c313d60ea35f67ef1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "820053bf94d9f49a6cd0600e25769f821f69b81962953a0c5f41c6865565bde3", "generation_seconds_batch": 24.447705950937234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.276336669921875, "target_z": 1.2656251097987778, "ce": 6.4515180587768555, "genuineness": 4.432034492492676, "identity": 0.806807816028595, "hypothesis": "De deur 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_S_MONO_reg/28f464be66d74182727ea12b7f64e1c06295dee817d94d7c313d60ea35f67ef1.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "2944c0c5826c24be5ff2bf7134419e60007f4af92d22283cbc682db553e74ae4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f53e95094cb0ceac7a4e7f90ca6dc930f4ede46800e86a1649f61b3b6343072c", "generation_seconds_batch": 24.067998295999132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 6.046441555023193, "genuineness": 4.570107460021973, "identity": 0.7222668528556824, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2944c0c5826c24be5ff2bf7134419e60007f4af92d22283cbc682db553e74ae4.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "2a1321e51d286cc96cb66a9402308b61822b769e02c7d9d93514e42710234980", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5161f80f5bfc5cc47057ae785c293ddf1e9ace741ede11ec50f22363b50e2cc5", "generation_seconds_batch": 24.05934224196244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.62030029296875, "target_z": 0.7500000234237393, "ce": 6.2767791748046875, "genuineness": 4.557555198669434, "identity": 0.7150000333786011, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2a1321e51d286cc96cb66a9402308b61822b769e02c7d9d93514e42710234980.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "2a28b5013e710ecb15a817dbd944246baba5e0381e02a3cb3841624547d2944f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "29679a2e9f91cc206325cbc45998213ed5010f4f68ff1acedb08352c599ec543", "generation_seconds_batch": 17.384654516004957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.308549880981445, "genuineness": 4.482243061065674, "identity": 0.7504522800445557, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2a28b5013e710ecb15a817dbd944246baba5e0381e02a3cb3841624547d2944f.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "2a2d312da5215b200fe81041e85ebaa9677f06ab557dcfed736b556380986b21", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "224f17faf38dbd9bd579044d3509f62eaf5ec6e1cae9aac70f165542f5596ccd", "generation_seconds_batch": 22.557529341080226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.186877250671387, "target_z": 1.1953126661621505, "ce": 6.229123115539551, "genuineness": 4.20609712600708, "identity": 0.8249484300613403, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2a2d312da5215b200fe81041e85ebaa9677f06ab557dcfed736b556380986b21.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "2b03f1bd8e243c1ea834e257fd00ddbd3de1e22d8595ecebb515d75a6be80ef3", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3efe351ebf0365bb7f1fb8b8921ac620a7c7abb977be3cdb174edbdd3508f396", "generation_seconds_batch": 17.029294381965883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.070269584655762, "genuineness": 4.808597087860107, "identity": 0.7636291980743408, "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_S_MONO_reg/2b03f1bd8e243c1ea834e257fd00ddbd3de1e22d8595ecebb515d75a6be80ef3.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "2b0e56f434f32cfebb54a1dca65105930e20673ca792c8d9e1e920f51a85f20f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7ed88e4f2dbc1d65f44239b9804eecb565c4ad5ecd20b590f94839b1ac27a62e", "generation_seconds_batch": 16.86037957202643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7185609340667725, "target_z": 0.04125969871636972, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.8095493316650391, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2b0e56f434f32cfebb54a1dca65105930e20673ca792c8d9e1e920f51a85f20f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "2b1b947906afd418bb51a09427dcd9f45c6bcc0bce000c9335967c8b7e70fbe6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "53fbb7ad6c72fffe21d5ede5328110cfab6bbd220a954c2172f7ee758cbf9219", "generation_seconds_batch": 23.812236011959612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8638288974761963, "target_z": 0.9414062657378248, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.7308125495910645, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2b1b947906afd418bb51a09427dcd9f45c6bcc0bce000c9335967c8b7e70fbe6.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "2b47ff65685dbcf47b6efba4eb2c3155de56bbf9e90a768722cfc3611f239042", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e5720284706f567f6b74bbd4fc69d33f0e9db5c40c0f842f087d0288a1a9c5f1", "generation_seconds_batch": 21.06101092195604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.125868320465088, "genuineness": 4.494794845581055, "identity": 0.6822647452354431, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2b47ff65685dbcf47b6efba4eb2c3155de56bbf9e90a768722cfc3611f239042.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "2b5c641a0762d42bf0ece03abf392112769d38185ac095485fc7a5f44c56df0c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ceb9339640e237370cebeea56128a5aa55029b23086f11af663cda5029df759", "generation_seconds_batch": 20.547841383027844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.227672576904297, "target_z": 0.4414063125853034, "ce": 6.157639026641846, "genuineness": 4.494794845581055, "identity": 0.5306546688079834, "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_S_MONO_reg/2b5c641a0762d42bf0ece03abf392112769d38185ac095485fc7a5f44c56df0c.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "2ba1272981fca906322bd07e7c6cfa7f2006cc8acfc6aae20c2b79ef29668997", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9dceecbd940edc7df35912dc542a2ce511b62feb29768a3cfc742f2281afe28", "generation_seconds_batch": 16.868734539020807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3270719051361084, "target_z": 0.5195312291382322, "ce": 6.316492557525635, "genuineness": 4.105680465698242, "identity": 0.22189590334892273, "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_S_MONO_reg/2ba1272981fca906322bd07e7c6cfa7f2006cc8acfc6aae20c2b79ef29668997.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "2bafd6867fc59f7e788a3e74b33fe102fb6e2e3db446382b1bda4d9d5e6a00b8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "16b189dbef4933acb8de1f4b8fd81654f00c32eb5f91618b27d2569ecea3c086", "generation_seconds_batch": 18.01490982202813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2962164878845215, "target_z": 1.2812500556313808, "ce": 6.23706579208374, "genuineness": 4.758388519287109, "identity": 0.7637733221054077, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2bafd6867fc59f7e788a3e74b33fe102fb6e2e3db446382b1bda4d9d5e6a00b8.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "2bd517d213251705bff0cca77d64b3ff0d96dd09b82b4f4b0d84221c688c72ec", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8681b0ff767b709d41c3d093cb2b642eed18fd8b74c1abbcfbbeff3e9ae15986", "generation_seconds_batch": 16.809002202004194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.299736976623535, "target_z": 0.4980467880759675, "ce": 6.078212261199951, "genuineness": 4.325342178344727, "identity": 0.6055330634117126, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2bd517d213251705bff0cca77d64b3ff0d96dd09b82b4f4b0d84221c688c72ec.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "2c27146cd07609c5f7d691488b3d8c736d8abe3aa911330c380b6e5b6a78957f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5c85c8078c17269be02781ba086f301dd43edd55d149bb72c2114735d9f43df4", "generation_seconds_batch": 16.987685942905955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9543240070343018, "target_z": 0.22656246413239922, "ce": 6.38003396987915, "genuineness": 4.281409740447998, "identity": 0.3096531331539154, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2c27146cd07609c5f7d691488b3d8c736d8abe3aa911330c380b6e5b6a78957f.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "2cb6803ee9b9d91cd18e52838f3c15bfa3efece2974c3becc15a1c5758305cf1", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6148682f47d11e93e34e7c482dd0645d6b1d25edba5693bd96c9e9b5f01156a", "generation_seconds_batch": 27.403635088005103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 5.919358730316162, "genuineness": 4.595211505889893, "identity": 0.6358540654182434, "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_S_MONO_reg/2cb6803ee9b9d91cd18e52838f3c15bfa3efece2974c3becc15a1c5758305cf1.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "2d2569edbb569b42a863f7a3fc157ea909fe88a5aba3dcfd166d7614d17c3df7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e4ff4fdf5716319a4c3ada38101996424de302d3511e2bffb7b4744be52a35e", "generation_seconds_batch": 25.545278916019015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.205295085906982, "genuineness": 4.356721878051758, "identity": 0.7058199644088745, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2d2569edbb569b42a863f7a3fc157ea909fe88a5aba3dcfd166d7614d17c3df7.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "2dabc3722e7028ef6b56baa9f103bb0f1e19e226a918fdf566375cb3361a4b5c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d8f6a512777be224a19d3b0924178cb701231d008f2928cd90be90eb904cac60", "generation_seconds_batch": 24.388625669060275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.073603391647339, "target_z": 0.3203125139078452, "ce": 6.221180438995361, "genuineness": 4.406930446624756, "identity": 0.47587940096855164, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2dabc3722e7028ef6b56baa9f103bb0f1e19e226a918fdf566375cb3361a4b5c.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "2dc6d8c385f14d0b0d4a90feb309657346530c3349177a348db10a6b0108ebf7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f0162a6caab1e830de6052c59ad0e831911169f3803c7c07f70c573db29917c9", "generation_seconds_batch": 16.003276851028204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.307192087173462, "target_z": 0.5039062833056293, "ce": 6.205295085906982, "genuineness": 4.093128204345703, "identity": 0.3458055257797241, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2dc6d8c385f14d0b0d4a90feb309657346530c3349177a348db10a6b0108ebf7.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "2eb981a797b3f5f082eee9bf29a918d8f9a81df5605aca3164c132569cd003f3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f6ff0e69aa165ad56ccff5a51311f9e4c807720d9a6bb8ba79bddf4634411cc6", "generation_seconds_batch": 18.915090066962875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.38567590713501, "target_z": 1.3515624992680082, "ce": 6.538887023925781, "genuineness": 4.645420074462891, "identity": 0.3603745698928833, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2eb981a797b3f5f082eee9bf29a918d8f9a81df5605aca3164c132569cd003f3.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "2f0217c20a562a2b40942c9476063d81e76ec36b51aff4d1dcd419208d197b95", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e4ad84c19c319b3b9a1374b2557caf9448aa7d0e3d2fb10fe942eac3cfe7146", "generation_seconds_batch": 19.913492696010508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.803361177444458, "target_z": 0.10791016864811201, "ce": 6.221180438995361, "genuineness": 3.9299514293670654, "identity": 0.056546129286289215, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2f0217c20a562a2b40942c9476063d81e76ec36b51aff4d1dcd419208d197b95.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "2fa75f695cffeeffcf4376bb2d09fad2071397b452ab52011daf988e0f47e5dc", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "adc89fd0a73a129a7278d5fb73b6a9e7913fbac2e51f3384e7b8406d6bd615bb", "generation_seconds_batch": 28.039855566923507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2972521781921387, "target_z": 0.49609381038932787, "ce": 6.340320587158203, "genuineness": 4.356721878051758, "identity": 0.3699836730957031, "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_S_MONO_reg/2fa75f695cffeeffcf4376bb2d09fad2071397b452ab52011daf988e0f47e5dc.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "2fe5416d5eb0464e7f034399123c1d8390a8a785c8266013895d1d007f6eed50", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f74573387d25b4b11c627cd1f5c22d7ff58153e08c8d65a31b840cab913e923", "generation_seconds_batch": 21.089074335992336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.9920125007629395, "target_z": 1.8281250336716253, "ce": 5.982900142669678, "genuineness": 4.68307638168335, "identity": 0.6502071619033813, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/2fe5416d5eb0464e7f034399123c1d8390a8a785c8266013895d1d007f6eed50.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "304098e26c4a95da68f5c178aa7a72b4b897a79235106917e114a21ccf369175", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "97579907108b2ca578e80ddb4423b0ee7b05ecabc0654e80a1dc14857c94fc39", "generation_seconds_batch": 17.332801937009208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.6375656723976135, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/304098e26c4a95da68f5c178aa7a72b4b897a79235106917e114a21ccf369175.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "30f07bd4886ea603cd5ec9fa422233bbdab8825db9a664b70592b333a7de0924", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1ae982b288354507e0e0990ddb5708e5bd7c83169aabe594340207e983333270", "generation_seconds_batch": 15.614239769987762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0885133743286133, "target_z": 0.3320313169772545, "ce": 6.165581703186035, "genuineness": 3.8671910762786865, "identity": 0.15580309927463531, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/30f07bd4886ea603cd5ec9fa422233bbdab8825db9a664b70592b333a7de0924.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "314be7ef6ed7ad919d2d1971ab076ad43f44b150530956ab05015ccbb928bd5c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ce6f4d981f2a3e8ee9d0e30e8befc2a6299a5ddb9ad4e40f8cbf876d384a9c2d", "generation_seconds_batch": 27.92381970596034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 6.109982967376709, "genuineness": 4.570107460021973, "identity": 0.7085052728652954, "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": 0.120000000000001, "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_S_MONO_reg/314be7ef6ed7ad919d2d1971ab076ad43f44b150530956ab05015ccbb928bd5c.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3158fc93dd8d6195b9e21595ef52db096008dacbfd2cda61766178efa3e7ff08", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "270425a10169478749ad180fd57b0b84e0128f092cc02aea99677d542c60444f", "generation_seconds_batch": 22.96240999503061, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6352100372314453, "target_z": 0.7617186391032343, "ce": 6.435632705688477, "genuineness": 4.457139015197754, "identity": 0.4109523594379425, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3158fc93dd8d6195b9e21595ef52db096008dacbfd2cda61766178efa3e7ff08.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "31d74ae7bb92365bebcb66b762a7a110751197501d31276fbd4e432cf9a63f52", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2192a221f5cc9e1dbf260e13e8c8eb13b845cc9d5f10ec328e6bbd2d16bcd5d5", "generation_seconds_batch": 22.91530083399266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7246696949005127, "target_z": 0.832031270129776, "ce": 6.149696350097656, "genuineness": 4.055472373962402, "identity": 0.41227811574935913, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/31d74ae7bb92365bebcb66b762a7a110751197501d31276fbd4e432cf9a63f52.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3202c83af6d14f3d51789a1c10a86f2860159e3d1b3d9dce5b98ffe49cf90cbd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e3f7c443aa89b61932b8bb56604e3dc03cb3ee7329ae2fb253f01fada213923e", "generation_seconds_batch": 22.28933443292044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9880781173706055, "target_z": 1.0390624582764645, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.5530514121055603, "hypothesis": "I found the little box. It was beside the old window. We can look inside to get", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3202c83af6d14f3d51789a1c10a86f2860159e3d1b3d9dce5b98ffe49cf90cbd.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "320cea6e0f64e501ce6a08b297e3e5ee728d47bc95cd4fd01693ebcf76c9fa27", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "85f216b0d9292a2ebd6e037a431e402311fb6c7f02dea53bd74879c08afe7e95", "generation_seconds_batch": 16.474709944915958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 6.403861999511719, "genuineness": 4.218649387359619, "identity": 0.4751601219177246, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/320cea6e0f64e501ce6a08b297e3e5ee728d47bc95cd4fd01693ebcf76c9fa27.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "3225d3557c33f761b2935471f25693033404f9da33ae179b8dc1111b40e00ef1", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a1d6a0137b23624c585271322c8786e2d6903abadfef53697244fa6d6779e33", "generation_seconds_batch": 27.970281027024612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087477684020996, "target_z": 1.1171875622193075, "ce": 6.538887023925781, "genuineness": 4.1998209953308105, "identity": 0.3184095323085785, "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_S_MONO_reg/3225d3557c33f761b2935471f25693033404f9da33ae179b8dc1111b40e00ef1.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "3241c418771f823e542938a26b59901ede81ceba9baba5d90da7309046ed4129", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "981a3abffd7974b87d3d71280ae7df8e3bedfa887595eb114627ee5b8be5e44a", "generation_seconds_batch": 23.866501386044547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087477684020996, "target_z": 1.1171875622193075, "ce": 5.824047088623047, "genuineness": 4.457139015197754, "identity": 0.5672849416732788, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3241c418771f823e542938a26b59901ede81ceba9baba5d90da7309046ed4129.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "32a45ce58a99cc398d847ebb8cd48acdb0bce6da14866bb41a664847c36c70aa", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fa79f28d79450c4b4352285456d1e22c9978e044a65876983b1f4535f592e415", "generation_seconds_batch": 23.991533047985286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 5.7406487464904785, "genuineness": 3.8169827461242676, "identity": 0.687605619430542, "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_S_MONO_reg/32a45ce58a99cc398d847ebb8cd48acdb0bce6da14866bb41a664847c36c70aa.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "32ad481d12148f284186ae23440969c82bbb6c7f7f49e84c3e929012794619b5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0591c17c2913344defda20b46a97421a0f4fc3c6ede395b1befd16b098c0dfb9", "generation_seconds_batch": 26.017455643042922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.272402286529541, "target_z": 0.47656253440361707, "ce": 6.133810997009277, "genuineness": 4.469690799713135, "identity": 0.7491462826728821, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/32ad481d12148f284186ae23440969c82bbb6c7f7f49e84c3e929012794619b5.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "32cf36ed1289a966caaebdf7b0dd21d6c72094972f22d22e15f00b867d5ada88", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "551c869ee7d717779d535c12551b071b12bcfa10ed158e0aebd5f4c29eb4198c", "generation_seconds_batch": 24.186107353074476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.355855941772461, "target_z": 1.3281248931291896, "ce": 5.542081832885742, "genuineness": 4.15588903427124, "identity": 0.5636257529258728, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/32cf36ed1289a966caaebdf7b0dd21d6c72094972f22d22e15f00b867d5ada88.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "32df7fa8e2ef3dcf447a4973f39a2f9e6b1166bbbbe4bdb3c28feeb33fb56edb", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f891c10168e92000bce1603c6de862f690a64ef5e40b3dddc4136024f7e5074c", "generation_seconds_batch": 20.497960462933406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 6.054384231567383, "genuineness": 4.406930446624756, "identity": 0.6866389513015747, "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_S_MONO_reg/32df7fa8e2ef3dcf447a4973f39a2f9e6b1166bbbbe4bdb3c28feeb33fb56edb.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "32e23070a38cda1829d875f1efe1a0de201c724721aaa3278269db4ee0bfe49b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "52ce169e29f13c96cc4b3bfe5ef02ec375e27ec17e4af287878641e5fbf14201", "generation_seconds_batch": 17.95903439295944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8389790058135986, "target_z": 0.921874989752114, "ce": 5.911416053771973, "genuineness": 4.570107460021973, "identity": 0.7950524687767029, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/32e23070a38cda1829d875f1efe1a0de201c724721aaa3278269db4ee0bfe49b.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "32fcaa4c2f6d2dee92251411986e9a6131756d3252b01d8375683b315b809360", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8a65e337ff9db4cbb7babfe34f4723093bf15d084b86591579948118c9dab6e1", "generation_seconds_batch": 16.537170646945015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.181467056274414, "genuineness": 4.419482707977295, "identity": 0.6336033344268799, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/32fcaa4c2f6d2dee92251411986e9a6131756d3252b01d8375683b315b809360.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3366dcfa93686b6980c824df9adffc21313ecb242f0476e7c41dffb5294370bc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "a69fac3cd656f0ae810f11b2bce62b777c55f84d8d22ffacea1a422b3a4b04e3", "generation_seconds_batch": 22.373950949055143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824069023132324, "target_z": 0.9101561866827047, "ce": 5.6334228515625, "genuineness": 4.068024158477783, "identity": 0.7625359296798706, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3366dcfa93686b6980c824df9adffc21313ecb242f0476e7c41dffb5294370bc.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "337ce9624e2b171faa63aa99c8e849d2aa664fa666b5262560d074f0898f2d02", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f9d4f44c3fcf28ed0c847e0b9c5e2299e5ee5f926c98011e2863fa7ff3523d64", "generation_seconds_batch": 22.7423298220383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.09409761428833, "genuineness": 4.3504462242126465, "identity": 0.6577849388122559, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/337ce9624e2b171faa63aa99c8e849d2aa664fa666b5262560d074f0898f2d02.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "33979e6c9dded452683fb3a628258525f003a68cf5efa99586cb5f9aef80cd6c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c46eeabdd86d6219e12056b915a7015fba0432445f4d2616f84aa115118bd851", "generation_seconds_batch": 19.913492696010508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1034231185913086, "target_z": 0.3437499326567496, "ce": 6.173524379730225, "genuineness": 4.124508857727051, "identity": 0.04571738839149475, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/33979e6c9dded452683fb3a628258525f003a68cf5efa99586cb5f9aef80cd6c.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "33cdd09efdaa9a2f2a05f574b2dcc26286410ba604f8dddd67464aa15d01777e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "74702ce7f5dbb40924a57f9421e80d8deb3398e61cdf64baa7ea1bb03afe23e5", "generation_seconds_batch": 23.882434190018103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 6.070269584655762, "genuineness": 4.65797233581543, "identity": 0.7027480006217957, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/33cdd09efdaa9a2f2a05f574b2dcc26286410ba604f8dddd67464aa15d01777e.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4be7abb23e851a29390cb7dbd73cd9f27dd7cc045987a925295c82f1ebf1ad3", "generation_seconds_batch": 16.535689910058863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2077927589416504, "target_z": 0.4257813667527005, "ce": 6.022613525390625, "genuineness": 4.18726921081543, "identity": 0.49423423409461975, "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_S_MONO_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "3400f47ed4d7fb41040611a4c9702b39879dae75b191a50acc5e49df2f2eaf34", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5130dde4e884a73b115af08739768c4122aceef064988b85500e9304a4298b9c", "generation_seconds_batch": 20.4519708189182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.7763037085533142, "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_S_MONO_reg/3400f47ed4d7fb41040611a4c9702b39879dae75b191a50acc5e49df2f2eaf34.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "345413fbca073bd8646a653b8e50eb3923e8bf19ca08479636ecc0f8c2816b45", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "305fb89ef4694872541aa2cbb14ae8c72fcc5ca393fc5193b9f165e22df0d2fc", "generation_seconds_batch": 29.258746451931074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.832973480224609, "target_z": 1.7031250922309735, "ce": 5.998785495758057, "genuineness": 4.582659721374512, "identity": 0.8165849447250366, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/345413fbca073bd8646a653b8e50eb3923e8bf19ca08479636ecc0f8c2816b45.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "f651583e8a713638330e15778007c373c40db27588334d1caf0ad6b8c93705be", "generation_seconds_batch": 17.713009259081446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 6.403861999511719, "genuineness": 4.218649387359619, "identity": 0.3089841604232788, "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_S_MONO_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3485278309709a5011626de1b7160a224e414e3f8657facf11991841bd9693e7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bd3adfdfea0247da8f4d408e1d15a252a4bcaa19ec4698047a1e1535cb566b7c", "generation_seconds_batch": 24.46639031090308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.501020908355713, "target_z": 0.6562499736482933, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.38211214542388916, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3485278309709a5011626de1b7160a224e414e3f8657facf11991841bd9693e7.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "34ba8945fd4dd295c0ed4cd561a16d0ed1196aeb4a9e73e59faa691c83a5566f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "682e1eaf6938b58ff642627a70688ac619bab45d184a2eae630b97e16f1d74c2", "generation_seconds_batch": 22.738490099902265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 5.998785495758057, "genuineness": 4.632867813110352, "identity": 0.6761536002159119, "hypothesis": "De deur 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_S_MONO_reg/34ba8945fd4dd295c0ed4cd561a16d0ed1196aeb4a9e73e59faa691c83a5566f.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "35374e23361f9feaa7ee35655528b842eda06ec838d65078117b7d8b43c29d43", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "472a5d940bf9ec79713a283432aaa52fd0ed1e6d406f4e4d44dd537e210d991e", "generation_seconds_batch": 17.674420837080106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6501200199127197, "target_z": 0.7734374421726437, "ce": 6.292664527893066, "genuineness": 4.287685871124268, "identity": 0.36203470826148987, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/35374e23361f9feaa7ee35655528b842eda06ec838d65078117b7d8b43c29d43.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "35885cf01e28214cbdadfbb0237b90ecf8f50dcc9577a13c535d10a90052a1fb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "21d18cc915576e6e7782aa5e7af664f43bec5984541f053bae91f8b622d133e8", "generation_seconds_batch": 23.882434190018103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.205295085906982, "genuineness": 4.632867813110352, "identity": 0.793948769569397, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/35885cf01e28214cbdadfbb0237b90ecf8f50dcc9577a13c535d10a90052a1fb.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "3590ea4cad3744ea267645703bb8f299562d4e507e99f532df547615a1b9428e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ed757a20de3ed002a81724bcfceee14e739cf783b342a45f2a262d49cc70ad15", "generation_seconds_batch": 28.180371741997078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.38797664642334, "genuineness": 4.15588903427124, "identity": 0.511218249797821, "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_S_MONO_reg/3590ea4cad3744ea267645703bb8f299562d4e507e99f532df547615a1b9428e.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "35b57fee1179f3f6cbed1799eac8ea5d43d4e78c2862d912c78d703a55a853a0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eee90756c1b9df7398f9ba0693e8c07666c9462fc9cd5c487b9540463d20ac4b", "generation_seconds_batch": 24.473477113060653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.999053716659546, "target_z": 0.2617186859507129, "ce": 6.1020402908325195, "genuineness": 3.9299514293670654, "identity": 0.20796550810337067, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/35b57fee1179f3f6cbed1799eac8ea5d43d4e78c2862d912c78d703a55a853a0.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "36710430ff545ac054865d064c34c4e82165f97f2ba0763c7e6227f9d59147d4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ae13cf842929c3d29865870ed513b43fc54833a90968eb6a4001e28c65e48bc", "generation_seconds_batch": 18.14616575895343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 5.855817794799805, "genuineness": 4.18099308013916, "identity": 0.4100184440612793, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/36710430ff545ac054865d064c34c4e82165f97f2ba0763c7e6227f9d59147d4.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "367303db87a8c43bfa082df9555a17765d3487624b79c6dd88062d56d0740291", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "ab08a1168d9f04005d1e1a3dc2459a8880fb6b8d186e3d21aaadb2244673df56", "generation_seconds_batch": 21.869271656963974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.6441144943237305, "target_z": 1.554687544651503, "ce": 6.348263263702393, "genuineness": 4.607763767242432, "identity": 0.2902436852455139, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "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_S_MONO_reg/367303db87a8c43bfa082df9555a17765d3487624b79c6dd88062d56d0740291.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "36beebc74bdb7fa9b1356b05b3fedec2a6127ece689c4cd43b5eb0243cfce7ab", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "348679bb15990377286b809e34f87391a3925fafcfa7b4cdfd41bc0c6ab12903", "generation_seconds_batch": 15.991079178988002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.280892848968506, "target_z": -0.30273440299868837, "ce": 5.935244083404541, "genuineness": 4.256305694580078, "identity": 0.6086850166320801, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/36beebc74bdb7fa9b1356b05b3fedec2a6127ece689c4cd43b5eb0243cfce7ab.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "36daabe87d22b8de1a37efc3c1553fc4b686857f050b336ce11014c7b5c86957", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bf5502f26f4ab0f0fc2a2e297b8d2cf23c0d820870a953773d14a2288f88d391", "generation_seconds_batch": 24.145741206943057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.213237762451172, "genuineness": 4.256305694580078, "identity": 0.6407302021980286, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/36daabe87d22b8de1a37efc3c1553fc4b686857f050b336ce11014c7b5c86957.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "37162b6c8612bf4f8c31688850eaeb887f37e68c535dfceda9b429683d07e8c9", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "026094dfa97ec67361319bb0bae58e3c03bfe04e61cf115a5bd696bb7a08550f", "generation_seconds_batch": 15.853977181017399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6451501846313477, "target_z": 0.7695312994094501, "ce": 6.419747352600098, "genuineness": 4.2751336097717285, "identity": 0.4948371946811676, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/37162b6c8612bf4f8c31688850eaeb887f37e68c535dfceda9b429683d07e8c9.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "373ad7677643d0a405d5cec7fee294b16986db0c4870fedf774b3d368d8b113a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "27957425ad128180a28c0c467b957a476d2c1e848e0f3795079da13fca22f490", "generation_seconds_batch": 23.678326043998823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.157639026641846, "genuineness": 3.8044304847717285, "identity": 0.48956099152565, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/373ad7677643d0a405d5cec7fee294b16986db0c4870fedf774b3d368d8b113a.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "378e6405fa03e94f9a86c964e09d6b659f6923a949f407c062859a9319f4df4c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6cba0144381a2b951acf683bcd44bc7ca64f0624c548e8eebc44b76c16d49968", "generation_seconds_batch": 23.938277384033427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 6.181467056274414, "genuineness": 4.519899368286133, "identity": 0.6389145851135254, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineingehen.", "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_S_MONO_reg/378e6405fa03e94f9a86c964e09d6b659f6923a949f407c062859a9319f4df4c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "37d144b47a9c70de0748f8345ad9341c40e86d63cb9ae4b9d8017b93fcaf2912", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cda53238da55216e911902ed515990df04127726bb97a09a972f1c6b6b5aa884", "generation_seconds_batch": 20.20499425707385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.753453731536865, "target_z": 1.6406249341207333, "ce": 5.959072113037109, "genuineness": 4.325342178344727, "identity": 0.6717516183853149, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/37d144b47a9c70de0748f8345ad9341c40e86d63cb9ae4b9d8017b93fcaf2912.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "37dc38276d6dd94da7eee1307241db429de1c1ae43e202428414403b86f8c69e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ee487bf06b9a47920f3d14266bca84633a94205dbeb9e7357115a1be8a7fc11", "generation_seconds_batch": 16.978754018084146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4950151443481445, "target_z": 1.4374998887372383, "ce": 6.173524379730225, "genuineness": 4.645420074462891, "identity": 0.6559529304504395, "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_S_MONO_reg/37dc38276d6dd94da7eee1307241db429de1c1ae43e202428414403b86f8c69e.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "37f7147a4d56ffe8f07fe46ba8ef0fb0e3e834c7d818990576c056775a626b35", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6a5f2cb4efe20376204986112b1572f13560d2841b1ae8c9903c572f6091b038", "generation_seconds_batch": 16.977997447014786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.38797664642334, "genuineness": 4.2751336097717285, "identity": 0.6649100184440613, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/37f7147a4d56ffe8f07fe46ba8ef0fb0e3e834c7d818990576c056775a626b35.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3857eed2e8d55e2fcdce08ddc73c9f3f691f104d0ebddc186fc0a74eba0916d9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "05afe10a6ed348572a723007c13ec5bf70895d0f2da87d8db39b147a3d34de68", "generation_seconds_batch": 24.418124006013386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.693814277648926, "target_z": 1.5937500966229246, "ce": 6.395919322967529, "genuineness": 4.557555198669434, "identity": 0.35159748792648315, "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_S_MONO_reg/3857eed2e8d55e2fcdce08ddc73c9f3f691f104d0ebddc186fc0a74eba0916d9.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "387b084709ba5f2b569590558b2702baf2ccccac7556f642b9d1e6e3ffb2866c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ebf43dc1503733a48ea21f092d73105631134658721169a7e582edf929b3bac0", "generation_seconds_batch": 25.27702296501957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9582583904266357, "target_z": 1.0156250395275601, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.1626964807510376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/387b084709ba5f2b569590558b2702baf2ccccac7556f642b9d1e6e3ffb2866c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3889ad99f3847ef3b8a9e56dd33c8bbe9eca85fc34c4a0989cfa5aaac4dc1bfd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fb096d939a4b0738927231999a9a713a7890a2e1f82c9043bcd4acbb8abb90b9", "generation_seconds_batch": 22.812258829013444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.117925643920898, "genuineness": 3.973883628845215, "identity": 0.7367290258407593, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3889ad99f3847ef3b8a9e56dd33c8bbe9eca85fc34c4a0989cfa5aaac4dc1bfd.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "398663eb4e2da81679ab63d5b9799cd2fa57ced122d5092f983975b1a0b84bf2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e75b7a654bb64c78962d37cdf72b5ae323dc6847ebe85880e14840a970e35951", "generation_seconds_batch": 24.023969881003723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.365796089172363, "target_z": 1.3359375534354052, "ce": 6.316492557525635, "genuineness": 4.400654315948486, "identity": 0.5130800008773804, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/398663eb4e2da81679ab63d5b9799cd2fa57ced122d5092f983975b1a0b84bf2.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "39d33d0d814f3d7c3b7732982b07d702a38bb1d20f03157ba1c8971e0a437da6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2f0832d94fa443e6f6506a2615976076e2da77abe4c7645e999178641c468820", "generation_seconds_batch": 23.60147459502332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 5.990842819213867, "genuineness": 4.068024158477783, "identity": 0.47862404584884644, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/39d33d0d814f3d7c3b7732982b07d702a38bb1d20f03157ba1c8971e0a437da6.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "39ed536eb60f8d7ec2bcd8de5f672d2cdf821c629cfc97ae57c52bd2c8c98cc7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "de555ed0b4d712e810361bd3bb5b54c599cbe10230b01f56704d4a0e1c7c4ac7", "generation_seconds_batch": 23.794210357940756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.4515180587768555, "genuineness": 4.6705241203308105, "identity": 0.6023411750793457, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/39ed536eb60f8d7ec2bcd8de5f672d2cdf821c629cfc97ae57c52bd2c8c98cc7.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3a517d640b88bab4de6e0c680cba56538291adfa7643c1b04e3741c4bba341db", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a2b99de7809fa10cb4c27d49330afc5ebdbd9b680f0b3bde663803644269ece4", "generation_seconds_batch": 17.133255761000328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.202822685241699, "target_z": 0.42187503659959263, "ce": 6.356205940246582, "genuineness": 4.369274139404297, "identity": 0.23789940774440765, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3a517d640b88bab4de6e0c680cba56538291adfa7643c1b04e3741c4bba341db.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "3a902acfb4ac411a87e65194b47f4001e74ceaaec13f649677a8b59977d35075", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "529704fe867e1e09d523f29c336d0bc5cc63776fab7029949d4077dc59897a25", "generation_seconds_batch": 17.036316379904747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0139636993408203, "target_z": 0.2734374890201222, "ce": 6.4912309646606445, "genuineness": 4.444586753845215, "identity": 0.3536544442176819, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3a902acfb4ac411a87e65194b47f4001e74ceaaec13f649677a8b59977d35075.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3aefd2067da4eef861f6eff006983b9df2a2b32d9f28bc17417310bfae2342a2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83f980b99fb1736d6702c5c0523e6da34707db97ebeac59d6c2ed6bf985acf74", "generation_seconds_batch": 15.194587405072525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 5.967014789581299, "genuineness": 4.118232727050781, "identity": 0.5414641499519348, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3aefd2067da4eef861f6eff006983b9df2a2b32d9f28bc17417310bfae2342a2.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "3b211c46b3f7027599b63fe11d2e9c807cd02d7ada14402037b93f133e15718c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f83696780b059ba0704510386aa6db02770d1d525a9a208f395875a403933bfe", "generation_seconds_batch": 15.195250277989544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.284721851348877, "genuineness": 4.3127899169921875, "identity": 0.7188436985015869, "hypothesis": "De deur 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_S_MONO_reg/3b211c46b3f7027599b63fe11d2e9c807cd02d7ada14402037b93f133e15718c.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "3b49408f3729e172b1f0a04c2430539c03e077318606148194ba624064fba0f3", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec731dfa0e13f58c4a80aa76faa275f8be7af70038a6e55cbda774cee9d96843", "generation_seconds_batch": 17.0018320819363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.037777900695801, "target_z": 1.0781250102478859, "ce": 5.967014789581299, "genuineness": 4.532451152801514, "identity": 0.8342080116271973, "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_S_MONO_reg/3b49408f3729e172b1f0a04c2430539c03e077318606148194ba624064fba0f3.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3c4299961d57706329ec8c8634cfa3a9781d936e23505a5401279ded1cc2fb92", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a79950392b162cbf204715ef1c098fec72dcb7b803377520ba0aee348662628f", "generation_seconds_batch": 18.44772389996797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.42647123336792, "target_z": 0.597656145691161, "ce": 6.268836498260498, "genuineness": 4.444586753845215, "identity": 0.7271517515182495, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3c4299961d57706329ec8c8634cfa3a9781d936e23505a5401279ded1cc2fb92.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3c4a565fd9f030b6938a3aa2287f4452f642f3e2644c7ae2e58014dec10aadf0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "10c538c6b80c3f5557449c5ae1120319895aa2fbd46c8dbac150f5519395b087", "generation_seconds_batch": 22.991529899067245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.127237319946289, "target_z": 1.1484374538845132, "ce": 6.165581703186035, "genuineness": 4.20609712600708, "identity": 0.64243483543396, "hypothesis": "Ik 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_S_MONO_reg/3c4a565fd9f030b6938a3aa2287f4452f642f3e2644c7ae2e58014dec10aadf0.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "2c65b8e9cdc55c0ab168e090c0a88052d06ea47cc5d31fcca5794054737b0d68", "generation_seconds_batch": 16.44418358593248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.24500846862793, "genuineness": 4.18099308013916, "identity": 0.7050732970237732, "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_S_MONO_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3c65e3fa17f76f2e94c199b15677a93110bd5b3ac4fee3956e2948ad1d2f3e6e", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b0218d098f523d93a63182cc5a8122b4574c854528b02e66fdf493b5e62cdec9", "generation_seconds_batch": 16.862392148002982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.4605168402194977, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3c65e3fa17f76f2e94c199b15677a93110bd5b3ac4fee3956e2948ad1d2f3e6e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3c69140512d3f2116f63dc5ab11dd66f6cb0b8f2030ada03f4adddad4c7b2912", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "332b4749fd588b869ac7aea918d6173e006f12ee3dd5f6d5f62c1b20251b06a6", "generation_seconds_batch": 15.194587405072525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.998018264770508, "target_z": 1.0468751185826801, "ce": 5.927301406860352, "genuineness": 4.331617832183838, "identity": 0.7757938504219055, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3c69140512d3f2116f63dc5ab11dd66f6cb0b8f2030ada03f4adddad4c7b2912.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "3c8e3e7dd6c4bd27c510ca4f0bd1e0f5e304f06756e11a4bdb0b33e31ba7dfdc", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "013c8e02c82835b0207c659fc0f95232d60b57993b6caf97b54932102af29265", "generation_seconds_batch": 16.420581302954815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.105908155441284, "target_z": 0.3457030977333035, "ce": 6.197352409362793, "genuineness": 4.036643981933594, "identity": 0.3378112316131592, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3c8e3e7dd6c4bd27c510ca4f0bd1e0f5e304f06756e11a4bdb0b33e31ba7dfdc.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "3cb891b278c508056adcfb77f5e0bd3076a7f5cbb0f7afdb5633d191669fb791", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0e9755f120fc71848eb06fdf1f557bb7645cb6d0b424affe10b8dade0ee09aeb", "generation_seconds_batch": 22.873072589049116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4861111640930176, "target_z": 0.6445313579687982, "ce": 6.268836498260498, "genuineness": 4.300237655639648, "identity": 0.3510627746582031, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3cb891b278c508056adcfb77f5e0bd3076a7f5cbb0f7afdb5633d191669fb791.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "3cb999d7a9de6988f5d104598eeed0e4c021d0d7160cfd9dd648a53f01d378c6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "390b05d297e53bd02781aae8cbda52ede9666c89bfe2d3a5134ff1597318e439", "generation_seconds_batch": 20.746464248979464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 6.109982967376709, "genuineness": 3.9173994064331055, "identity": 0.4645351469516754, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3cb999d7a9de6988f5d104598eeed0e4c021d0d7160cfd9dd648a53f01d378c6.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "3d166ef078e04b39c194b026c323a8d06322fe3b06b0012faf58ab844c69b761", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60d026e75a4385e01cee083fc65a7488ea19e8a4ff4dc07b4685ce5d6f645e85", "generation_seconds_batch": 22.983800269081257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.324435234069824, "genuineness": 4.024092197418213, "identity": 0.3029068112373352, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3d166ef078e04b39c194b026c323a8d06322fe3b06b0012faf58ab844c69b761.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "3d37f2bac5a77a2a04ce457ac0ebc316f17135ddb5dd368259162b59622af115", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c1cd4e82ea933bc5bc8f26c84a91d47be722e73ed19bfaa3369fc521973bfbab", "generation_seconds_batch": 17.002526212949306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 6.427689552307129, "genuineness": 4.356721878051758, "identity": 0.6821306943893433, "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": 0.120000000000001, "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_S_MONO_reg/3d37f2bac5a77a2a04ce457ac0ebc316f17135ddb5dd368259162b59622af115.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3d58aeaa8f6dcf5aaad56d8602675cbc1f07b3c08effcf213b111625863aa0a4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "69b241894e7fea2494399ebd56fd3cf9d5d29c7d0a6e226128c28f07a4150b0f", "generation_seconds_batch": 24.442791369045153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.763393402099609, "target_z": 1.6484372196471204, "ce": 6.197352409362793, "genuineness": 4.231201171875, "identity": 0.5743918418884277, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3d58aeaa8f6dcf5aaad56d8602675cbc1f07b3c08effcf213b111625863aa0a4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "3d9fd0c02ab783fe9cec1de55cb432009b64283fb1b0ea8eddd9d19b6eed8f99", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c7a207ddb619bbeee94960324bfc37bb7faf95c989994d9d6259f2a844fc08ef", "generation_seconds_batch": 17.52979721489828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1034231185913086, "target_z": 0.3437499326567496, "ce": 6.038498878479004, "genuineness": 4.545003414154053, "identity": 0.7583390474319458, "hypothesis": "He encontrado la cajita. Estaba junto. a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3d9fd0c02ab783fe9cec1de55cb432009b64283fb1b0ea8eddd9d19b6eed8f99.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3dae7371542fd43d320adebd4af56891d0e339bf1e1fb25cab215d5e430f5cb0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "21bec5c14967c983a689354dbdc46182620a2136d1fc7e3b3bb611729cdf63f7", "generation_seconds_batch": 23.840342373005114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.952252388000488, "target_z": 1.7968747672265908, "ce": 5.784333229064941, "genuineness": 4.457139015197754, "identity": 0.6600949764251709, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás ligado.", "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_S_MONO_reg/3dae7371542fd43d320adebd4af56891d0e339bf1e1fb25cab215d5e430f5cb0.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "3dbcad48a1d71b50325f84de90999bef88c8f88004b5f618203a070a2d07ff79", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ecf268fb2dcf6838b764ce2a769839e065aea24ba12bd3652380cc034084df92", "generation_seconds_batch": 21.06101092195604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.141753673553467, "genuineness": 4.080576419830322, "identity": 0.5469314455986023, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3dbcad48a1d71b50325f84de90999bef88c8f88004b5f618203a070a2d07ff79.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "3dc44a5e3220326180cc8af58a98ef7f4653f551e9b77ad615de589866ccf6ff", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ab6e997fc7e93c8c792728315f96ebadd95757d09f5ae91563f6916afb77c423", "generation_seconds_batch": 24.7436652519973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.683874130249023, "target_z": 1.585937436316709, "ce": 6.483288764953613, "genuineness": 4.582659721374512, "identity": 0.6287765502929688, "hypothesis": "Ik 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_S_MONO_reg/3dc44a5e3220326180cc8af58a98ef7f4653f551e9b77ad615de589866ccf6ff.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "3e775a9081cf30687f1a7aec5cce643321f8b5c820b330d39ee29405a9fc2e8a", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63c04657b7d1a3031095429162fcae62da0a5a3374694a321c132199b2c0c7b5", "generation_seconds_batch": 27.804643949959427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.7246347665786743, "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_S_MONO_reg/3e775a9081cf30687f1a7aec5cce643321f8b5c820b330d39ee29405a9fc2e8a.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3e98d0823927dbbba52cc0618e62483fc9bcdb39aca96c160048f2a550dc0599", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f0d01e00eae25b3a98ff4655bdaee522878e2e804ab8b943532c9a120b5de1ab", "generation_seconds_batch": 25.185939830960706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.77933931350708, "target_z": 0.874999964864391, "ce": 5.903473377227783, "genuineness": 4.042920112609863, "identity": 0.48340103030204773, "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_S_MONO_reg/3e98d0823927dbbba52cc0618e62483fc9bcdb39aca96c160048f2a550dc0599.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "3ea8e8208f92ccb8bea559e355dfdc917da5c91cc6a636f5ee873d1baeddbfac", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f8fe8bba6a797dcaa3d33adc8322d13d710c65c2647ba0ff515aad9e1a8c0d56", "generation_seconds_batch": 21.15967216505669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.046441555023193, "genuineness": 4.243753433227539, "identity": 0.6589429378509521, "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_S_MONO_reg/3ea8e8208f92ccb8bea559e355dfdc917da5c91cc6a636f5ee873d1baeddbfac.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "3f557eebbd1d81fee45232c241c4ac2e802f255f47e1df276c20cc4c74b3b95b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d23295ea076c53ac5f6e0fb9730b79cb690b5e90b4d843f0b672bf9e3f67f7b5", "generation_seconds_batch": 17.82468868000433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 6.292664527893066, "genuineness": 3.9927117824554443, "identity": 0.7014281749725342, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3f557eebbd1d81fee45232c241c4ac2e802f255f47e1df276c20cc4c74b3b95b.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "3fcc44d9083cf750eae5837ede30fe52a97ea2ec782636d8810ad37c1fcae685", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b7c19f9e751b052e2edb96bff6cbd4e9189b77e043f22469ceb517257c44fe9a", "generation_seconds_batch": 16.408988456008956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5293915271759033, "target_z": -0.10742183053149496, "ce": 6.046441555023193, "genuineness": 4.2249250411987305, "identity": 0.5416070818901062, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3fcc44d9083cf750eae5837ede30fe52a97ea2ec782636d8810ad37c1fcae685.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "3fd1b87c970112a917cea44940d4ced97b01d31f0bab3dba95bac8ced20c18b7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6bea097087c028002fb6028e4cf3a6d52fb4f3d3b8c2c81c130ff3790495b600", "generation_seconds_batch": 19.73513227608055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.227672576904297, "target_z": 0.4414063125853034, "ce": 5.919358730316162, "genuineness": 4.6705241203308105, "identity": 0.7031870484352112, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3fd1b87c970112a917cea44940d4ced97b01d31f0bab3dba95bac8ced20c18b7.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "3ffd3ff3c9c74826e317718d5acc2f4ffe1165ee4951aed33fdea1a7839adcd4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "588e4c2f17a72e096f05bb518d70d265a469fb384cb8e47bcb85b91a73747f18", "generation_seconds_batch": 18.44803942902945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.307192087173462, "target_z": 0.5039062833056293, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.7656354308128357, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/3ffd3ff3c9c74826e317718d5acc2f4ffe1165ee4951aed33fdea1a7839adcd4.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "4015f7b354dfe10bd91c1e54ac935ec755abba4d34f60be61cf8ce94d21c2b17", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c37b0095863304f04bb038f38fd1f07076ada9cadcd383a8490bed0018406242", "generation_seconds_batch": 20.181037642993033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9035887718200684, "target_z": 0.9726563447929449, "ce": 5.855817794799805, "genuineness": 3.8546390533447266, "identity": 0.6393474340438843, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4015f7b354dfe10bd91c1e54ac935ec755abba4d34f60be61cf8ce94d21c2b17.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "402bb3f3b10e610392a0d83681c69eab74a9c7bcec8e4360da7d3cadcb9edd55", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "5597bf57ea06ab4b9a146cc16a1250d61ea60a61bec2d9b83625e42030cdcc0d", "generation_seconds_batch": 27.782143476069905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.268836498260498, "genuineness": 4.444586753845215, "identity": 0.5784476399421692, "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": 0.120000000000001, "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_S_MONO_reg/402bb3f3b10e610392a0d83681c69eab74a9c7bcec8e4360da7d3cadcb9edd55.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "403306bd5112424a0292fb62c7f83927396be2ca6929067ef73248d53666d952", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eaa7da0acce2cd501330c82a7822a692f96a16c657cd2b427f0f5d5c5ceec307", "generation_seconds_batch": 23.765001379069872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.332377910614014, "genuineness": 4.017816066741943, "identity": 0.3551786243915558, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/403306bd5112424a0292fb62c7f83927396be2ca6929067ef73248d53666d952.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "4048d99186ea76be7ed2b70f5064c2c3db0581baf562a87b2e85ed053e184fce", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7c05db280f901fd48a4bcf45f44969e745a70dd02cdfd453be31e72648d45670", "generation_seconds_batch": 17.17114224296529, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.912079334259033, "target_z": 0.19335940739063948, "ce": 6.252951145172119, "genuineness": 4.2374773025512695, "identity": 0.4042987525463104, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4048d99186ea76be7ed2b70f5064c2c3db0581baf562a87b2e85ed053e184fce.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4054ca85b05bfbc54862345b6acc7913ad8e7f5667dd1c9386b474abcbc681ed", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3515320399c7761bdc51fbba7f5159050b356b5465d90e9c802aab71fe3a9495", "generation_seconds_batch": 17.035671227960847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.739579677581787, "target_z": 0.8437500731991853, "ce": 6.054384231567383, "genuineness": 4.469690799713135, "identity": 0.5555651783943176, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4054ca85b05bfbc54862345b6acc7913ad8e7f5667dd1c9386b474abcbc681ed.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "405912b2d4dd51ec35fd80eca675c4a08ac55536ec5843afaa2bdb92c2b7e56a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2ed34e3371d033d3383b75891aa81acc407cc4097fadf486e86f69c060b3f5df", "generation_seconds_batch": 22.785985962022096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.634174346923828, "target_z": 1.5468748843452873, "ce": 5.597680568695068, "genuineness": 4.3881025314331055, "identity": 0.8433356285095215, "hypothesis": "J'ai trouvé la petite boîte. Elle était prêt de la vie fenêtre. We've regarded ensemble the.", "wer": 0.42105263157894735, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/405912b2d4dd51ec35fd80eca675c4a08ac55536ec5843afaa2bdb92c2b7e56a.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "40a2e5c2a646e5afdc91ec2a73b327d9ffc133553d924370c6a16258889f56ac", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a12c6fe0bce34c7c96b4557ea7330c3f4589a082ba95b2303b56e69dadc52e8f", "generation_seconds_batch": 21.9684867199976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.777268886566162, "target_z": 0.0874024038190814, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.7731480598449707, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/40a2e5c2a646e5afdc91ec2a73b327d9ffc133553d924370c6a16258889f56ac.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "40d9496d6fb31ee1ebe655600974536574c60aca3430d90109dc7878046d668b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2777b99c6fdfbf8ad5dbb47a0af00757d868a4f84a93a9940c6b743884c2d707", "generation_seconds_batch": 23.137947648065165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.7369307279586792, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle connaît.", "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_S_MONO_reg/40d9496d6fb31ee1ebe655600974536574c60aca3430d90109dc7878046d668b.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "416a084a701dae5598f299886db333a3bb7e55162f326ddd9f4b1ae4dd78a512", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2f653e5e29f4fb74bdcaeda2aa0af4af55d037213a16f8aff9dbe4cf74b7427a", "generation_seconds_batch": 24.01446759304963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.5575869083404541, "hypothesis": "Ich habe die kleine Schacht 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_S_MONO_reg/416a084a701dae5598f299886db333a3bb7e55162f326ddd9f4b1ae4dd78a512.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4172c10d80c690b39435418d3723855f121a59722180b2d752fdb07f2a0d0c8d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d72d11779a3aba7eef4d4bdc02c62b92f9ed092cd789f4838b62878cc2674d57", "generation_seconds_batch": 23.67650514608249, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7843093872070312, "target_z": 0.878906295017499, "ce": 6.109982967376709, "genuineness": 4.7834930419921875, "identity": 0.6681715250015259, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4172c10d80c690b39435418d3723855f121a59722180b2d752fdb07f2a0d0c8d.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "417713093d3b3c74fbee9dd762fe8077579c42803177467d41e235b5d7d34320", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "138e08829904dfb765c4af52ef8558a7a9dc8d741bee2351bd91ce539b272023", "generation_seconds_batch": 15.95428179006558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.93837833404541, "target_z": 0.9999999063050429, "ce": 6.3641486167907715, "genuineness": 4.193545341491699, "identity": 0.3853118419647217, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/417713093d3b3c74fbee9dd762fe8077579c42803177467d41e235b5d7d34320.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "4182797c4f4a273082b2000cd42114d0339778b77787dffd9f48067bb9049363", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6bfd6f53205498b0d494c6cdf3ea7bdfcaf6ff2e6842e4388ec69baa95a88960", "generation_seconds_batch": 21.977638877928257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.814129114151001, "target_z": 0.9023437137664033, "ce": 6.070269584655762, "genuineness": 4.607763767242432, "identity": 0.6414399743080139, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4182797c4f4a273082b2000cd42114d0339778b77787dffd9f48067bb9049363.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "4199f829a66d702ab7ef40e9e93ed5c776dbffd5ff1632a34811d6ac14e1b933", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e07749b0b915ee277d1451ecd6b413dfc4a4c62ad9c9591557132f0a321ee524", "generation_seconds_batch": 17.15684340405278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.23706579208374, "genuineness": 4.394378662109375, "identity": 0.45951026678085327, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4199f829a66d702ab7ef40e9e93ed5c776dbffd5ff1632a34811d6ac14e1b933.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "420bbec987b2491eb47875346182d6e01561099f63aab27dd318e34480b92e88", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f988e6113e3551591676a2329425d10a96315be4d4820697476e79a5c0f2d0d2", "generation_seconds_batch": 16.88931248593144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2127625942230225, "target_z": 0.4296875095158941, "ce": 6.213237762451172, "genuineness": 4.65797233581543, "identity": 0.7962619066238403, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/420bbec987b2491eb47875346182d6e01561099f63aab27dd318e34480b92e88.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "420ea0921ec0e68c38b4e7f6f31e08c28f4b96709b819e3b13d181a44e89db66", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9f2b0b80396855613a59441e82a89b89014fc3804e6550708dd2087c08fb39a8", "generation_seconds_batch": 22.669119808939286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.928438663482666, "target_z": 0.9921876207786557, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.6750575304031372, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/420ea0921ec0e68c38b4e7f6f31e08c28f4b96709b819e3b13d181a44e89db66.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "4221b9c339ef5648929892919312c9aac5853d148780be86cee316efdf6c02af", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a98f613eb76e564f2576e2069150a75d0433da43513f764b29106b183c1b3ab2", "generation_seconds_batch": 20.599265680997632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.999053716659546, "target_z": 0.2617186859507129, "ce": 6.324435234069824, "genuineness": 4.381826400756836, "identity": 0.5033289194107056, "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_S_MONO_reg/4221b9c339ef5648929892919312c9aac5853d148780be86cee316efdf6c02af.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4263cd8f9593ceef73623ef4d6f86683d3de2d64fafcc66be646867e3de72241", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "984da87e70a3900c55e27c6552456bb640d79e625de58293e09a5e61a8e61a99", "generation_seconds_batch": 23.709162737010047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 6.078212261199951, "genuineness": 4.331617832183838, "identity": 0.8802074193954468, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4263cd8f9593ceef73623ef4d6f86683d3de2d64fafcc66be646867e3de72241.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "42654775357d69eb94f3f7b2528f7e4baee2ef9ce406adbfe95c1bd003b97ad5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "515609491d361218cda6d7a9e85e51d9c35b3f007674dedf6227152bcb66a079", "generation_seconds_batch": 24.527191777015105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.706138551235199, "hypothesis": "J'ai trouvé la petite de 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_S_MONO_reg/42654775357d69eb94f3f7b2528f7e4baee2ef9ce406adbfe95c1bd003b97ad5.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "42b1b1368d28a5242991a98b89eda1fdaa0b836dfc992c41c43e7066b99d3dd0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "488bb49bfd2ddaa0c95704558b7354ba7855343ffeef9af7f8198ba2d8725230", "generation_seconds_batch": 24.34885942400433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614294528961182, "target_z": 1.5312499385126843, "ce": 6.181467056274414, "genuineness": 4.507347106933594, "identity": 0.4211598038673401, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/42b1b1368d28a5242991a98b89eda1fdaa0b836dfc992c41c43e7066b99d3dd0.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "fca9e0baa33e727d18ec4849f619694f7b851b6614a5a81c3d159600ba88f45c", "generation_seconds_batch": 16.535689910058863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.799633741378784, "target_z": 0.10498051472823824, "ce": 6.046441555023193, "genuineness": 4.268857479095459, "identity": 0.5668278932571411, "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_S_MONO_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "43930faec61e4a1371299874cb8a2dd03ade3a0e19755679c9ee185eadfb8030", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d623206ab567b37f170ea42c4e0fbe20fbcd1c47563922dbaf3f62ddd1ae70a6", "generation_seconds_batch": 26.90478867595084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614294528961182, "target_z": 1.5312499385126843, "ce": 6.062326908111572, "genuineness": 4.720732688903809, "identity": 0.7604435682296753, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/43930faec61e4a1371299874cb8a2dd03ade3a0e19755679c9ee185eadfb8030.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "43cb65ea596557a9f4ade17e500800c922110528567d526af867ea1c874e06d6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ade693f31e72f92709ab86f8f692390b7e3e35782d1cd231ea4a37602837622d", "generation_seconds_batch": 17.50185189000331, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.4525051414966583, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/43cb65ea596557a9f4ade17e500800c922110528567d526af867ea1c874e06d6.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "a5459728ae5eda113adddc2398a57a27fd8632e93c0a744a07f41c057f77aeea", "generation_seconds_batch": 16.697768932092004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5159308910369873, "target_z": 0.6679687767177026, "ce": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.6501604318618774, "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_S_MONO_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "43db01f7795d639ec5cc1bc6282a781277f81ee9d502e76690e5d62ca2a47cc1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "20e0e88436b751a433f302e1720d373ae783b566b7f69b1790fa18386b9b95aa", "generation_seconds_batch": 23.934990959940478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.09409761428833, "genuineness": 4.20609712600708, "identity": 0.5935210585594177, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/43db01f7795d639ec5cc1bc6282a781277f81ee9d502e76690e5d62ca2a47cc1.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "43f2203fc88cc0a0783494cb3830748f13a318e6a741fe7b3074ff12f655c83d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e0fdf90efe43c9f15eed99f1d7829c82491392423dd045b429a4ef67111c5727", "generation_seconds_batch": 21.590709432959557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.395919322967529, "genuineness": 4.65797233581543, "identity": 0.387650728225708, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/43f2203fc88cc0a0783494cb3830748f13a318e6a741fe7b3074ff12f655c83d.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "44060916caeac8afd10212bded475bc2eb0db6baaa64cbca80f11d5125635c2f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b48705d9a73154ce82efe59bef1a745ec1e3cf4c5f20ee3413b21fe8e189147b", "generation_seconds_batch": 20.06638449395541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 5.8875885009765625, "genuineness": 4.6705241203308105, "identity": 0.6527090072631836, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/44060916caeac8afd10212bded475bc2eb0db6baaa64cbca80f11d5125635c2f.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "446de62905c41152a312010248e7c756c6f09f3987d5e6779bac9185345c8163", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "76673c63522c725c83f6c403433554298066365b3e21abfe734c5372eb56bc98", "generation_seconds_batch": 24.34437019797042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 6.022613525390625, "genuineness": 4.055472373962402, "identity": 0.5929257869720459, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/446de62905c41152a312010248e7c756c6f09f3987d5e6779bac9185345c8163.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "448fd2d52aba2d8a18511bd97e38601af5a033b71878d00cf7dc2a7eeac9e6df", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7f6da704812e77fd99f696979f1d3d96f362445f7523e8b523fe1d50625862bf", "generation_seconds_batch": 21.108577296021394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.206757068634033, "target_z": 1.2109376119947535, "ce": 5.847874641418457, "genuineness": 4.15588903427124, "identity": 0.7025017738342285, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/448fd2d52aba2d8a18511bd97e38601af5a033b71878d00cf7dc2a7eeac9e6df.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "449e7a02096d268bd6e0f33ac997162155676d4945e9394b1454cf888604ef06", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e11b13b2fce4281a158f70b2c940e57dcc92957a54df4d4328d8b5d840a11420", "generation_seconds_batch": 27.56565720506478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.514895439147949, "target_z": 1.4531252093496698, "ce": 6.125868320465088, "genuineness": 4.2249250411987305, "identity": 0.709487795829773, "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_S_MONO_reg/449e7a02096d268bd6e0f33ac997162155676d4945e9394b1454cf888604ef06.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "44afe3850f44e1f7ce2e591858ccf4299eb48d530c97f3f3628f7a7023f82519", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c26e2917e5abbdda20a4f2c0aa95a2de0782004c4b774381ae28292d00648891", "generation_seconds_batch": 22.243789308005944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 6.252951145172119, "genuineness": 4.250029563903809, "identity": 0.7363201975822449, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/44afe3850f44e1f7ce2e591858ccf4299eb48d530c97f3f3628f7a7023f82519.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "44c1f6d42a1c93f8614965521476b32c131e129c04864200719fa3feaab84561", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1ef7ab795664285c06d720edeca13b5fa435c6cd02bc56c157980b21845d4a41", "generation_seconds_batch": 22.870237127994187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 5.998785495758057, "genuineness": 4.582659721374512, "identity": 0.6081686019897461, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/44c1f6d42a1c93f8614965521476b32c131e129c04864200719fa3feaab84561.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "456366300c096c3c0cca487adf8fc01d5f12f912d30d3332c598502561b54685", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6a001fd9327c48cd915e25ccc20009a0726e8d69d8dde4aa9c9b76f896880a2f", "generation_seconds_batch": 22.968772301916033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.038498878479004, "genuineness": 4.65797233581543, "identity": 0.6817891597747803, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/456366300c096c3c0cca487adf8fc01d5f12f912d30d3332c598502561b54685.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "45b1355811ee6f350c163d4ef20e2d69a415edd6ed9ae8e453414f3fa69e4da6", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23957553cb60d4f91e61317978a38824e95ee1c6df48ca6c82f0ce9dfcf92b9e", "generation_seconds_batch": 27.804643949959427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.514895439147949, "target_z": 1.4531252093496698, "ce": 6.1894097328186035, "genuineness": 4.595211505889893, "identity": 0.7198781967163086, "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_S_MONO_reg/45b1355811ee6f350c163d4ef20e2d69a415edd6ed9ae8e453414f3fa69e4da6.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "45f0688d84013482c8c90828b05f5b9052e8c2c2be0462c5ff20c80f96c68983", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d091b0b404d3b34e56153de6f7d00de5e101e1f106c2345de2dc08bbd2ccfdb0", "generation_seconds_batch": 21.929005658021197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.046441555023193, "genuineness": 4.262581825256348, "identity": 0.7338535785675049, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/45f0688d84013482c8c90828b05f5b9052e8c2c2be0462c5ff20c80f96c68983.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "bfdcbc451a8c8f49df2e88174256482fe7b6fd2f5ca9af8cdd47912f8887be3e", "generation_seconds_batch": 15.251756650046445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.501020908355713, "target_z": 0.6562499736482933, "ce": 6.316492557525635, "genuineness": 4.218649387359619, "identity": 0.6292387247085571, "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_S_MONO_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "46dd879d0301df59a71ffc71aa67bd897c68e22654f98025e794fccbd05948fe", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cc8c0e795bf9ce48b359e60e483af5891268559cd8c7fca3113da8f0304312d6", "generation_seconds_batch": 16.68167663703207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3221020698547363, "target_z": 0.5156250863750386, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.8472542762756348, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/46dd879d0301df59a71ffc71aa67bd897c68e22654f98025e794fccbd05948fe.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "474c7e868a8acc5cf993a7db4990d753dbea0add6e476dcac141ed3bd9662cf8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2905fec03e41211b9eb0340b10a2ccee4426d996e7861990c0d1aa67d774df25", "generation_seconds_batch": 22.785985962022096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.514895439147949, "target_z": 1.4531252093496698, "ce": 6.260893821716309, "genuineness": 4.7834930419921875, "identity": 0.7803956270217896, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/474c7e868a8acc5cf993a7db4990d753dbea0add6e476dcac141ed3bd9662cf8.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "8c8bb9732bcad51893457437f1c3492e0bab197f6880b39966fe824ba318a2b7", "generation_seconds_batch": 15.281586281955242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.8312159180641174, "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_S_MONO_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "481f7df40c99b7ff03c77259fdbb32c7c17a177248228aad46bd96b3a737e225", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eafd80f7a68582f0651d9f4c64ed4cbb60b4bdb2eca87c806fd332830296d8f6", "generation_seconds_batch": 15.641998409992084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.594414710998535, "target_z": 1.5156249926800816, "ce": 6.268836498260498, "genuineness": 4.507347106933594, "identity": 0.26492565870285034, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/481f7df40c99b7ff03c77259fdbb32c7c17a177248228aad46bd96b3a737e225.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "4846394152f20d4bd97e091870d652bc378fd4fb42a25ea01de0f1a2206e09f2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f36810ff3a63f135cebda64a5acb312ebd10f01f9c8b8ea9ab45af8cb3f165a3", "generation_seconds_batch": 23.771782980998978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.312161922454834, "target_z": 0.5078124260688229, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.3630651831626892, "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_S_MONO_reg/4846394152f20d4bd97e091870d652bc378fd4fb42a25ea01de0f1a2206e09f2.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "4858881d5ed992cf0c58f5416b1b1d9c6e98818b7769d5659a6668305456f0df", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "80dc5b414cb773d4ea87f79f22f97c8c7faf4fc677bc3024e225a5ef4ffd4294", "generation_seconds_batch": 19.548522931989282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.037777900695801, "target_z": 1.0781250102478859, "ce": 6.38797664642334, "genuineness": 4.482243061065674, "identity": 0.45799434185028076, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4858881d5ed992cf0c58f5416b1b1d9c6e98818b7769d5659a6668305456f0df.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "487231f1301dcf7c24c178a6002ead73343a596763f8028fe206464d58213902", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fe78165503c9587c8bd72e76d247aeb9b777637a9c080f421a4f684e2eaaeb31", "generation_seconds_batch": 16.90658159099985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 6.141753673553467, "genuineness": 4.105680465698242, "identity": 0.38181012868881226, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/487231f1301dcf7c24c178a6002ead73343a596763f8028fe206464d58213902.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "48d1391ff3e36369042db3e77ef59a4cf926895140025177d3e5689479a1696b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e0368ffb519c092c9139b7a6f3be5b360cf22e546949f97e62358cfc91ff0283", "generation_seconds_batch": 16.537170646945015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.038813591003418, "target_z": 0.29296876500583297, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.6504735946655273, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/48d1391ff3e36369042db3e77ef59a4cf926895140025177d3e5689479a1696b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "49187b5ec0a8f73148c885c56d3929814fd400b2a8a15214f1ec15c29a82276c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "469341fb3b83f7038a423c0dc2f7e200453561ce119e2b8c58be476fe6a3ae6f", "generation_seconds_batch": 20.746464248979464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.284721851348877, "genuineness": 4.2374773025512695, "identity": 0.4159066379070282, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/49187b5ec0a8f73148c885c56d3929814fd400b2a8a15214f1ec15c29a82276c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "4927bf5034a355c0e61b6d1ea1c7c32489568b3b7144b91140bce3f58056f086", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dd0fd5f7e00fcc073e5f4fe0daebed79305f75bc89a57545a9a3efd1ef9df159", "generation_seconds_batch": 18.18009413301479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6849098205566406, "target_z": 0.8007811910746558, "ce": 5.990842819213867, "genuineness": 3.8922953605651855, "identity": 0.4815348982810974, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4927bf5034a355c0e61b6d1ea1c7c32489568b3b7144b91140bce3f58056f086.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "0d4ed87fc70f0b321465274b2b3b592e4de70b270b4b770614baaeff3ae69289", "generation_seconds_batch": 16.51340415002778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.615330219268799, "target_z": 0.7460936932706315, "ce": 6.181467056274414, "genuineness": 4.068024158477783, "identity": 0.6724568605422974, "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_S_MONO_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "4a7617680fead93b511f0023f717eb49c929358ef18519e6a36ec8434fa4ee19", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "68de4c81ec47bb752d2db9c56acf196964f20b9a20233e2dcf3cab48a3ffb41c", "generation_seconds_batch": 26.170985282980837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.3641486167907715, "genuineness": 4.18099308013916, "identity": 0.6711999177932739, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "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_S_MONO_reg/4a7617680fead93b511f0023f717eb49c929358ef18519e6a36ec8434fa4ee19.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "4acc6858cd6131c5117bdcca92589999c6758bf6a26596ff54d17c97ddad6db1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "af0ec7d281da636255447dd6a428d544be6345208e1bdcbf2e2f8d258e3c198c", "generation_seconds_batch": 25.545278916019015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 6.09409761428833, "genuineness": 4.356721878051758, "identity": 0.21825851500034332, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4acc6858cd6131c5117bdcca92589999c6758bf6a26596ff54d17c97ddad6db1.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "4b1cdf53319cf81601c839eb0c7f22d677461e651c36c8dfddc2b9a32b989c0b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "65040d84a83eb49fe0f16d4ad773a45bc00380910a3abb1b7848f1f5b83bd308", "generation_seconds_batch": 17.036316379904747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 6.38003396987915, "genuineness": 4.400654315948486, "identity": 0.40728527307510376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4b1cdf53319cf81601c839eb0c7f22d677461e651c36c8dfddc2b9a32b989c0b.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "4bafeff880af991771a90ccbdb1314835de8a131a511f7842aaebcdf1bea7ff8", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c632b3e4000b5a44d8c873675d3afbe3cb723ac40c6aa259ce8cd2fa607401cb", "generation_seconds_batch": 28.039855566923507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.332377910614014, "genuineness": 4.049196243286133, "identity": 0.45697009563446045, "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_S_MONO_reg/4bafeff880af991771a90ccbdb1314835de8a131a511f7842aaebcdf1bea7ff8.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "4bc61f27e84140f08d35177373706601e07a9468ee58bc9cf1a483835c338792", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3dcab3331ea68e798765c7a638b815f2af93ebe3ce1296cb89f0dba491bee7b2", "generation_seconds_batch": 17.82468868000433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.538884699344635, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4bc61f27e84140f08d35177373706601e07a9468ee58bc9cf1a483835c338792.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "4cd573c4627ea9a1346525f3cc7e0b9a5d8b87b39cdca54214bd957d6259c69b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b8756bc0e40f59286f0ec6fd0c5e6fed5bcadcd87b2dec9782cf681463a66feb", "generation_seconds_batch": 23.920740696950816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9965686798095703, "target_z": 0.25976552087415894, "ce": 6.141753673553467, "genuineness": 4.18099308013916, "identity": 0.5843942165374756, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4cd573c4627ea9a1346525f3cc7e0b9a5d8b87b39cdca54214bd957d6259c69b.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "4cf8aef9a9b496ad1cdbfae8c71dd8cb338acffb5611b8937667bc4d4272ee1c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2e27eddf80fe9e55e6374858aafd3438fcb137b70a0dfc8b1dc7e1dec79818ba", "generation_seconds_batch": 29.08080533402972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994083881378174, "target_z": 0.2578125431875193, "ce": 6.165581703186035, "genuineness": 4.432034492492676, "identity": 0.6999763250350952, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4cf8aef9a9b496ad1cdbfae8c71dd8cb338acffb5611b8937667bc4d4272ee1c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "4d46b6e5def3d32ecda25b61b636c6f6ba9ca58d502026f1cc2bbe345ecaa298", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e90b44f4bf9dc003e02287d89fbbdaed6f49adfeccbe6ba919319f40141dfd06", "generation_seconds_batch": 22.557529341080226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.213237762451172, "genuineness": 4.149612903594971, "identity": 0.678375244140625, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4d46b6e5def3d32ecda25b61b636c6f6ba9ca58d502026f1cc2bbe345ecaa298.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "4d49304d8bbdfaf10152906cd9b8d576f0026c7565cba5aea0111462888c1548", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "27b3e30f6532e47d90a9eff638139f01690c258d9daafd8d2b24c906bd4fb1da", "generation_seconds_batch": 18.363242020946927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 5.839932441711426, "genuineness": 4.15588903427124, "identity": 0.6332446336746216, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4d49304d8bbdfaf10152906cd9b8d576f0026c7565cba5aea0111462888c1548.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4d8c84a0549105ff848172aa281be377990bf0204768a36fda7961115fe20f66", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1e8781b6b3cc02c6c972c612fa531658ea6feca62978e31da7921e5e95c9aaaf", "generation_seconds_batch": 21.252716325921938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.021418571472168, "target_z": 0.27929679685986974, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.8450251817703247, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4d8c84a0549105ff848172aa281be377990bf0204768a36fda7961115fe20f66.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "4d8f457e86b6be30427fa4254b2a3638c41f054051797fe6378c867002c72781", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d3bf3f8735e21c25a996139ed6ea2a7370c1bc5d7b3ddc94efdac49cac814c75", "generation_seconds_batch": 17.376099015935324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.23761248588562, "target_z": 0.4492187855016049, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.40224745869636536, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4d8f457e86b6be30427fa4254b2a3638c41f054051797fe6378c867002c72781.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "4e05eddfae8ecc66d72c410f257bbd57c28dae07f14f3af19b0c79b083d9cc47", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3e20daa48e532ae6bcff9d3eaef2c9f300f8e52b2e7d527e0d370456d364a77a", "generation_seconds_batch": 25.39391974499449, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.1562654674053192, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4e05eddfae8ecc66d72c410f257bbd57c28dae07f14f3af19b0c79b083d9cc47.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "4e2b196aaf6108d1510766d78aaea9b1d759ad5e29c4ab3d97c45ed4e2138740", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9e7fd5e5c2d9a2e1c794f346ce0cc0b29971420e0079fc369129db7e4e78167c", "generation_seconds_batch": 28.152758267940953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.6833016276359558, "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_S_MONO_reg/4e2b196aaf6108d1510766d78aaea9b1d759ad5e29c4ab3d97c45ed4e2138740.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4e8d7c581e9a1e4a8054f07ea4d0aa63c027d750558075278210946eb75f1b71", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e3498fbb99797277a97a04cba9454b78c49dc7e3aa06ed6f9e9f26b09718970a", "generation_seconds_batch": 17.326142848003656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 6.062326908111572, "genuineness": 4.268857479095459, "identity": 0.8287497758865356, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4e8d7c581e9a1e4a8054f07ea4d0aa63c027d750558075278210946eb75f1b71.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "4eab3ded4787aec9102893559f357d663ae64216519ad854403182df8300ff2a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0e48ce3d6e0c8ebf2d1b25599549461ac54e8a1d0a7cac15bc99ee401c61b512", "generation_seconds_batch": 23.702739716041833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.051238536834717, "target_z": 0.30273440299868837, "ce": 6.09409761428833, "genuineness": 4.13078498840332, "identity": 0.6288988590240479, "hypothesis": "De deur 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_S_MONO_reg/4eab3ded4787aec9102893559f357d663ae64216519ad854403182df8300ff2a.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "4eb71655717883198ab48cb6f94324fd4f2b0f8eb337f3f452937f81247491f2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "470c60500d8f2a83a4e60c273b2580d55590dd75faee362436834eda1e4384c1", "generation_seconds_batch": 23.860926738008857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614294528961182, "target_z": 1.5312499385126843, "ce": 6.070269584655762, "genuineness": 4.883909225463867, "identity": 0.7463017106056213, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/4eb71655717883198ab48cb6f94324fd4f2b0f8eb337f3f452937f81247491f2.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "4f082526ddb376d907a819beb1f32a9359ad26e12ade687b9419e72619714451", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05dbcc344beb54efb67700a65e09d96389fe5924efd4ebc7f77f92541b02bdb3", "generation_seconds_batch": 20.45865741802845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3359761238098145, "target_z": 1.3124999472965866, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.7000044584274292, "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_S_MONO_reg/4f082526ddb376d907a819beb1f32a9359ad26e12ade687b9419e72619714451.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "4f63e37db746c8f224379b1183aeb25f17cf649269a726fb7ca0d1365b3d5643", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c171252882f6e75785a8fdaddda79ed2f57fc8e4053566b322d40dc62479d12c", "generation_seconds_batch": 24.81174633000046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 5.982900142669678, "genuineness": 4.482243061065674, "identity": 0.6693646311759949, "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_S_MONO_reg/4f63e37db746c8f224379b1183aeb25f17cf649269a726fb7ca0d1365b3d5643.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "4f6ebafaf034c6c03e2f72b1c5ffae53be6463ca711ac2b95e066f6b24ec89ae", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e7ffb71376f0be3662efa7fc1abefbbe401194de33fefb1c4879b26d27b3cd3f", "generation_seconds_batch": 27.775130321038887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.125868320465088, "genuineness": 4.557555198669434, "identity": 0.6073926687240601, "hypothesis": "La porte 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_S_MONO_reg/4f6ebafaf034c6c03e2f72b1c5ffae53be6463ca711ac2b95e066f6b24ec89ae.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "5034c44cb5f11c252df2994dfb092d489a898a60a24fb4f407d25c801c07cdbe", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2379c47576a66f1318265bb41cdea1ab5091f10e52971495a6e77111594dd7f2", "generation_seconds_batch": 24.520703764050268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.23706579208374, "genuineness": 4.457139015197754, "identity": 0.5628684759140015, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5034c44cb5f11c252df2994dfb092d489a898a60a24fb4f407d25c801c07cdbe.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "503e73c4f2fe60ab4ee0bef5e13a02f584831b4bd65ceda3d25da2961a4307ce", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bc52b2bbd93bcbae7895bd1c994b6828be1dda45b1f3fa755ef9ebb1d92add35", "generation_seconds_batch": 22.95910521398764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.375736236572266, "target_z": 1.343750213741621, "ce": 6.09409761428833, "genuineness": 4.231201171875, "identity": 0.7620297074317932, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/503e73c4f2fe60ab4ee0bef5e13a02f584831b4bd65ceda3d25da2961a4307ce.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "5086178d98508339859d95f87facf69da685a67afd340d06bd8c04532e2cb701", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b6bfaa5db2cc13c710228ca07d54421cc54fdef802e3236688f1a4a39dba14e2", "generation_seconds_batch": 21.926722283940762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.158092975616455, "target_z": 0.38671881478127895, "ce": 6.141753673553467, "genuineness": 4.1998209953308105, "identity": 0.8493181467056274, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5086178d98508339859d95f87facf69da685a67afd340d06bd8c04532e2cb701.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "50cdb33e4dcda4869bc3cf1a17a9cf361958a045ad6a094da36c4ac89697017c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "64c7a3bb2db7f805abd5bffa056a8578fdfc4df807cd5ec45785251c79b32355", "generation_seconds_batch": 16.60791674396023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6451501846313477, "target_z": 0.7695312994094501, "ce": 6.443574905395508, "genuineness": 4.243753433227539, "identity": 0.4185076951980591, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/50cdb33e4dcda4869bc3cf1a17a9cf361958a045ad6a094da36c4ac89697017c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "50f152e8bf552451fa9a8ca25e12e29eb6d1f561a99c4374665a572fd8ef10be", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8bae3e6f6c048d2bf7b4be0403bf3ea5e3713e717d2d3141cdbdffd75ac61c9b", "generation_seconds_batch": 21.105477695004083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.22663688659668, "target_z": 1.2265625578273565, "ce": 6.173524379730225, "genuineness": 4.7332844734191895, "identity": 0.495229572057724, "hypothesis": "Ik 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_S_MONO_reg/50f152e8bf552451fa9a8ca25e12e29eb6d1f561a99c4374665a572fd8ef10be.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "5151d547608f6cf7882aa014e714e982a0555a71546071997b6274b85703b581", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "adffd8b8ccd3028eba8f7f7cb6b35bda5f0888a47aa7d4ebc94a0dd47bbfb3fa", "generation_seconds_batch": 24.48224454803858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 5.998785495758057, "genuineness": 4.65797233581543, "identity": 0.5761497020721436, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5151d547608f6cf7882aa014e714e982a0555a71546071997b6274b85703b581.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "5186cfc34c02be7c5403198d499179dce2e0c10998ed9c544b0f0c522e72a73c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c6724bfe9ec1ae3866e992c697da65bd81e060fbf9b5bcaf2bcfeb88319522b6", "generation_seconds_batch": 23.98368084896356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 5.982900142669678, "genuineness": 4.557555198669434, "identity": 0.6433115601539612, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5186cfc34c02be7c5403198d499179dce2e0c10998ed9c544b0f0c522e72a73c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "51aaeddc57e8580051c60afb01a839e56b28f716bc164f01dd112db58b1bc262", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ab3f39913e3351b5892cf84caf62e3f8e25f8ca7c6d8f55d9ef9a3ed45586586", "generation_seconds_batch": 16.661556991981342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.070269584655762, "genuineness": 4.268857479095459, "identity": 0.14662159979343414, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/51aaeddc57e8580051c60afb01a839e56b28f716bc164f01dd112db58b1bc262.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "51ecf408e775754eb948b9afb67a51c23ed22565e7047b958d52f449aa803869", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fdc6a556d04cf835435fc476b61a3c0c93cdde4c36d4476c21a4ee846bf1d0c6", "generation_seconds_batch": 20.780079418909736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.372091293334961, "genuineness": 4.319066047668457, "identity": 0.694631814956665, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/51ecf408e775754eb948b9afb67a51c23ed22565e7047b958d52f449aa803869.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "52010c4d992ef848ddef53dde8215d8399aca1fe536491fb96cd230fc64eb17e", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "321a090467e76c5d32c3c302c6e11b3ab412d18a09e056715e2800b23c4ea10f", "generation_seconds_batch": 18.435334975016303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.5810544490814209, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/52010c4d992ef848ddef53dde8215d8399aca1fe536491fb96cd230fc64eb17e.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "5211896812e3c48d1805a12202208e2513558ae4a0742432ed5bd288af5050c4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "998db47388d3b49952f1638d2ef1069114e6d8d9b5e843c3bd2c93940a409536", "generation_seconds_batch": 17.292658876045607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8524396419525146, "target_z": 0.14648438250291648, "ce": 6.2767791748046875, "genuineness": 4.482243061065674, "identity": 0.5926364064216614, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5211896812e3c48d1805a12202208e2513558ae4a0742432ed5bd288af5050c4.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "521aad95b15952c46b4ddaf6536df08ac5ca436f1763ebae6ce0d4fe5a7c33ad", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e98460df7b14076a2780a14324723b822d8f00f3c098a94d99279b5b660bdb7b", "generation_seconds_batch": 17.649989911937155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.308549880981445, "genuineness": 4.256305694580078, "identity": 0.4666805863380432, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/521aad95b15952c46b4ddaf6536df08ac5ca436f1763ebae6ce0d4fe5a7c33ad.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "523b46e5a9b3dfb3774533b6222b96f2935e7e0e782360a3758922581439c3e3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "73f42eee44ac166716eaea35a23c505acfa2494f264428f48480d7550d344c52", "generation_seconds_batch": 25.76906446600333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1471171379089355, "target_z": 1.1640623997171162, "ce": 6.24500846862793, "genuineness": 4.15588903427124, "identity": 0.2630387246608734, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/523b46e5a9b3dfb3774533b6222b96f2935e7e0e782360a3758922581439c3e3.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "5273e560f5152e97300cb3962f61ef0b1d12099e2ad138c3677cdcef945dfc12", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ac9736eb330a49f58a6054e617a22d14447696e9518cd91b4eea556f95bd2a39", "generation_seconds_batch": 22.264705550973304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.654054641723633, "target_z": 1.5625002049577188, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.7425439357757568, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5273e560f5152e97300cb3962f61ef0b1d12099e2ad138c3677cdcef945dfc12.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "52bf21acc30e8c2dfe6fffb6b5b4a74ae1fbe5e6c4340c06035ba52b5193e163", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9003fa860680b4738e57e87708b0c06e02992f3c37acebfda0872ec7e1cbfca2", "generation_seconds_batch": 26.026475038030185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.100938320159912, "target_z": 0.3417969549701099, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.6229710578918457, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/52bf21acc30e8c2dfe6fffb6b5b4a74ae1fbe5e6c4340c06035ba52b5193e163.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "52f0384e49e2013915036bfdeb9e4ff2bbc7bac438710c67b4b66b12efc006b3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3f2b70b6bc3cdd26c2713749b0a947b06e2bcb863c7567a775247417380c72c8", "generation_seconds_batch": 24.01446759304963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087477684020996, "target_z": 1.1171875622193075, "ce": 6.4912309646606445, "genuineness": 4.287685871124268, "identity": 0.6126640439033508, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/52f0384e49e2013915036bfdeb9e4ff2bbc7bac438710c67b4b66b12efc006b3.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "52f3925022f3d921e1ac5323494134504bc0d43a8b9e986f76e4acca67cee875", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cc961f21768fdd80083973a5852a3780e6a50f4e67d6aa0752b6ad67eee6a781", "generation_seconds_batch": 22.873072589049116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6053903102874756, "target_z": 0.73828122035433, "ce": 6.324435234069824, "genuineness": 4.231201171875, "identity": 0.4376561641693115, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/52f3925022f3d921e1ac5323494134504bc0d43a8b9e986f76e4acca67cee875.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "534e331262a275bee5eecda520b8ef94a9baa3966df7ca028591495f3e3622bf", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7286a8aa2f68a6ecbebc1aebc6f7c3e7842256dd7848448d2e90e10995972bf0", "generation_seconds_batch": 27.978949897922575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.395919322967529, "genuineness": 4.331617832183838, "identity": 0.5249737501144409, "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_S_MONO_reg/534e331262a275bee5eecda520b8ef94a9baa3966df7ca028591495f3e3622bf.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "536bbc6c85ddd2cd6248391c1db075adf601ba3a5b930fc783fc585138aa677e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "345b0282d18d76fc8af0df263ee2812f4472e9b41a9e36c429913946dd016267", "generation_seconds_batch": 22.65118133602664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.445315361022949, "target_z": 1.398437336765817, "ce": 5.756534099578857, "genuineness": 4.218649387359619, "identity": 0.6514421701431274, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/536bbc6c85ddd2cd6248391c1db075adf601ba3a5b930fc783fc585138aa677e.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "53b8559b83abff59f5345ad33ee9108c5d2c5929c34c287a69fa14fa079352a9", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d0f3b7628fae3ac044a660c209392cecb1adac243ed09b245219fb291807e71f", "generation_seconds_batch": 17.044571025995538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 6.23706579208374, "genuineness": 4.293961524963379, "identity": 0.8017817139625549, "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": 0.120000000000001, "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_S_MONO_reg/53b8559b83abff59f5345ad33ee9108c5d2c5929c34c287a69fa14fa079352a9.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "53d21c70bb90f563acaa5cf2495df3b990be59ed7ca1549bd9877e1968a9f17e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ae0bc30dc07504ce2345038068f4ce9a274c70e3a8dbc2de8e58d8b6b97307fa", "generation_seconds_batch": 23.30289683502633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2301573753356934, "target_z": 0.44335929027194304, "ce": 6.109982967376709, "genuineness": 4.325342178344727, "identity": 0.49004077911376953, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/53d21c70bb90f563acaa5cf2495df3b990be59ed7ca1549bd9877e1968a9f17e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "53fb7e8fffea2f31845b32693c392735744f5049eebfb5be35585a857c79f109", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ae411e6750903b3609859bc1e8f549b6cafd76bf0e2ea2ae4d98139e8ad15748", "generation_seconds_batch": 22.373950949055143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 5.442798614501953, "genuineness": 4.118232727050781, "identity": 0.6459747552871704, "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_S_MONO_reg/53fb7e8fffea2f31845b32693c392735744f5049eebfb5be35585a857c79f109.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "5416324ae09d100a68a7f138901473f07b7356a299725d0c41d7b38d7f207d55", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6601e0d08342a3828908ef9989c684398609cbfc0a601a2b448adedc55c85e52", "generation_seconds_batch": 20.51666610792745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.897169351577759, "target_z": 0.18164060432123016, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.7625325918197632, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5416324ae09d100a68a7f138901473f07b7356a299725d0c41d7b38d7f207d55.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "545354e201b245c33a737bf24095350b88a2038328f7cbd2008996f713c15e5a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dfc51b45c167709f6cf344141f0cf4fba034afc554cb484b5fa3075af5ca33d6", "generation_seconds_batch": 23.069345214986242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.3641486167907715, "genuineness": 4.337893962860107, "identity": 0.4024029076099396, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/545354e201b245c33a737bf24095350b88a2038328f7cbd2008996f713c15e5a.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "f7d911f75ab6841d35174f10470163861d48e8c38ce8140769ffc3345541995e", "generation_seconds_batch": 15.251756650046445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3022220134735107, "target_z": 0.49999995315252144, "ce": 6.252951145172119, "genuineness": 4.105680465698242, "identity": 0.6556094884872437, "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_S_MONO_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "54681e32b31b72acd7b2437338dcd18586fbc8c5276ec18ba2aaf5564e465815", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1835021eb457e57d7924e09b9f45d1eac97ea6d35c581db34f72ced3cba4b89a", "generation_seconds_batch": 25.11648467509076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.962192535400391, "target_z": 1.8046874275328066, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.7079079747200012, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/54681e32b31b72acd7b2437338dcd18586fbc8c5276ec18ba2aaf5564e465815.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "55662757905ef4d776fc970c0677fe86d93ba2d067e6015cecc422e46b42fd4a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "24b863e640c347bce73fe8d7b2bcbfb3c1a9d6ced07092d611ad1afd580a15c4", "generation_seconds_batch": 28.32258645503316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.078212261199951, "genuineness": 4.632867813110352, "identity": 0.468511700630188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/55662757905ef4d776fc970c0677fe86d93ba2d067e6015cecc422e46b42fd4a.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "558ac341f41730bd9f04f3314aa033364c779b0a4ecc711e6107c3561ad97588", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3bd9fa844215c01d2be2e9bce86aadba6acb7f5b98840ad3c5c0ed3241d98694", "generation_seconds_batch": 17.37892849103082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.419747352600098, "genuineness": 4.293961524963379, "identity": 0.44141262769699097, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/558ac341f41730bd9f04f3314aa033364c779b0a4ecc711e6107c3561ad97588.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "55e268de1cc27b2e6663fde40ffc35cf6d309aad50b18e37eef41e606c3b4ac9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c61d95644c43beda21d4a5f651c6f08192d6f370bd3915660a247ec38b98bc43", "generation_seconds_batch": 18.389203201048076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.77933931350708, "target_z": 0.874999964864391, "ce": 6.23706579208374, "genuineness": 4.143336772918701, "identity": 0.3307298421859741, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/55e268de1cc27b2e6663fde40ffc35cf6d309aad50b18e37eef41e606c3b4ac9.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "55e86b7794485c4ee7ace74c8187aafbe18ded553704179d687d876210374b17", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "25557be3fa03c1db88f6c33b2d29bbda3b5f14de3958791503520f1e11247e03", "generation_seconds_batch": 17.92705337505322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.38003396987915, "genuineness": 4.118232727050781, "identity": 0.6444849967956543, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/55e86b7794485c4ee7ace74c8187aafbe18ded553704179d687d876210374b17.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "55ec949b454bb6182d414c05b491ad6965523dc06dc764a45441de024cbb9256", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cb31e4a61b32b7bf0f88bc909ad7b0318f47fa86d534a13b7e40737e5df80e95", "generation_seconds_batch": 21.318227363051847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.205295085906982, "genuineness": 3.9048471450805664, "identity": 0.25431257486343384, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/55ec949b454bb6182d414c05b491ad6965523dc06dc764a45441de024cbb9256.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "56b1af82a8e5e39125a5f7ea2d786fae238949963ac114b5d532638c48baba9f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3d68455ba5b73257cedba3ae99a349e5017faf4dc40c617d690587d025d88145", "generation_seconds_batch": 17.596382108982652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.016448736190796, "target_z": 0.2753906540966761, "ce": 6.229123115539551, "genuineness": 3.8295347690582275, "identity": 0.2954990267753601, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/56b1af82a8e5e39125a5f7ea2d786fae238949963ac114b5d532638c48baba9f.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "570cf0f1a40cf7a185f2d28e0a2f474d89c9569da5ace9df3960808571777637", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "abe4e4feacf0472d1015886c984d14ba48731945b0b451bf3445b6e37d6f5df4", "generation_seconds_batch": 18.939092115033418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 6.252951145172119, "genuineness": 4.632867813110352, "identity": 0.3564687967300415, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/570cf0f1a40cf7a185f2d28e0a2f474d89c9569da5ace9df3960808571777637.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "57618a4308d3fba223a62909d6449e8d096618009bd60595c89a8715860fad21", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5a09e55aa696fb721327a47ba0b730be8aadcc67ff2c84338750d58156f80d69", "generation_seconds_batch": 17.257368410006166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8561670780181885, "target_z": 0.14941403642279025, "ce": 6.1020402908325195, "genuineness": 3.9550557136535645, "identity": 0.48946499824523926, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/57618a4308d3fba223a62909d6449e8d096618009bd60595c89a8715860fad21.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "5786c25f105a0405ffbf86deb04795e15ea1ffbcf1000f91bd12a8fdb2b166a7", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af86225670dfaaa1fd2b256d6bcb46f8963d7956553f1a42a55e54d66cfc92b3", "generation_seconds_batch": 16.765714254928753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.505990982055664, "target_z": 0.6601563038014012, "ce": 6.292664527893066, "genuineness": 4.406930446624756, "identity": 0.5192118883132935, "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_S_MONO_reg/5786c25f105a0405ffbf86deb04795e15ea1ffbcf1000f91bd12a8fdb2b166a7.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "57894a95f4ad46d84afb5e710645d6c5c49dcb6a790e7ba5638db2ff477c316c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8b6d05fff84d8b1cdbd22f92a4ac606460896fa0cbb70f7f82e96326865e0ae", "generation_seconds_batch": 17.058715209015645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.157057285308838, "target_z": 1.1718750600233319, "ce": 6.141753673553467, "genuineness": 4.7081804275512695, "identity": 0.7670234441757202, "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_S_MONO_reg/57894a95f4ad46d84afb5e710645d6c5c49dcb6a790e7ba5638db2ff477c316c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "58084120ec8a81dc16b78d032bbed797e3d4784fdf67f619ba4174b9f5e7cf3e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "04793adfd7d04eaf5c744e3c16e1090e2790214f45fba34342f168176e86f779", "generation_seconds_batch": 23.770071857026778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.118333101272583, "target_z": 0.3554687357261589, "ce": 6.1894097328186035, "genuineness": 4.3127899169921875, "identity": 0.6638807654380798, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/58084120ec8a81dc16b78d032bbed797e3d4784fdf67f619ba4174b9f5e7cf3e.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "580ff3d97db988e891b1a4a2399f2419fa21141810d7b3309ec62438fc821fa6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bbb36f98353905bd70479fc04fcc3c99dfa071758688bda296617765ea4ed7f0", "generation_seconds_batch": 24.7436652519973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.574534893035889, "target_z": 1.5000000468474786, "ce": 6.09409761428833, "genuineness": 4.143336772918701, "identity": 0.6875746250152588, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/580ff3d97db988e891b1a4a2399f2419fa21141810d7b3309ec62438fc821fa6.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "58751736afc6720e144c513c59f3d10bc5596fdaaaf2119c5c1f86db62f8f3cd", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e52377f7f5950dfaf1d3b03e0126ddfa2feabf7d9c081410769fa974e75fba", "generation_seconds_batch": 16.927631431026384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.693814277648926, "target_z": 1.5937500966229246, "ce": 6.0305562019348145, "genuineness": 4.344170093536377, "identity": 0.8527717590332031, "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_S_MONO_reg/58751736afc6720e144c513c59f3d10bc5596fdaaaf2119c5c1f86db62f8f3cd.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "589bd4e5e90de2641ce37f9d8d4c8f80eeb86cdbdcfbd9212ccf998997a98c7d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "23ef1a0cb840de720b3fc82c279b64cfb8c733225d7bc998a10032ff04d79508", "generation_seconds_batch": 21.988214793032967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3767716884613037, "target_z": 0.5585937811096537, "ce": 6.229123115539551, "genuineness": 4.406930446624756, "identity": 0.7124571800231934, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/589bd4e5e90de2641ce37f9d8d4c8f80eeb86cdbdcfbd9212ccf998997a98c7d.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "58b2b45547cb3e4e73a110363a8c0dcb40f5fd855843ebc05e9f8141f96b3c4c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b7ed3f7a99c5ef097c981cf9ee2be1974d2e47c8b58a80075a4268b847680ae7", "generation_seconds_batch": 18.915090066962875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.123303174972534, "target_z": 0.35937506587926676, "ce": 6.284721851348877, "genuineness": 4.362998008728027, "identity": 0.4116419553756714, "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_S_MONO_reg/58b2b45547cb3e4e73a110363a8c0dcb40f5fd855843ebc05e9f8141f96b3c4c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "58c018f77f563eab8403c936b1373b99e4155d4dfee194da84c62f18a0435a93", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6dc7dd1596ce395489d96bc7e4065008deee7e65cfce5cb01479c2018073cbf1", "generation_seconds_batch": 17.015110059990548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227025032043457, "target_z": 0.4374999824321955, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.5099961161613464, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/58c018f77f563eab8403c936b1373b99e4155d4dfee194da84c62f18a0435a93.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "59151ba50519c74a89a16bbf06333bbb400a50ec9d3ef7eb58eb65632f5971e0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d76c1c8c97363322d562b73ea3b286d53fd842de678aea311d656a0b54fd78b3", "generation_seconds_batch": 23.75978673994541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.1894097328186035, "genuineness": 3.986435890197754, "identity": 0.3370567262172699, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/59151ba50519c74a89a16bbf06333bbb400a50ec9d3ef7eb58eb65632f5971e0.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "5a13070afe8c9bd0b64d7a5e75f6fb67a4c843a75422746ab549804682d014ef", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b7af1bd7a57ab2f33dc9c8917e26b3dd661cc432f9994a5a688d1165774db424", "generation_seconds_batch": 25.259155779029243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.205295085906982, "genuineness": 4.306513786315918, "identity": 0.7872834205627441, "hypothesis": "La porte 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_S_MONO_reg/5a13070afe8c9bd0b64d7a5e75f6fb67a4c843a75422746ab549804682d014ef.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "5a6970e48b4fba6928b06fc0f200c318235106714bba62277ccbf5b941d482f4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ba87061b51264943ec75d18352bf8d4ef12186c914c3943b1ac92a9e2bb3cd9e", "generation_seconds_batch": 16.357562349992804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.046441555023193, "genuineness": 4.306513786315918, "identity": 0.5438392162322998, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5a6970e48b4fba6928b06fc0f200c318235106714bba62277ccbf5b941d482f4.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "5abb9ce565b10f22d59d3185cd50dd1e1f95bc8072702622d2fb04f84b5fa421", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "87fc7fc83f962e85c1c5ae278a86dc217e4839377d1b0411ce50d84031834b48", "generation_seconds_batch": 22.739073859062046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.157057285308838, "target_z": 1.1718750600233319, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.6256037354469299, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5abb9ce565b10f22d59d3185cd50dd1e1f95bc8072702622d2fb04f84b5fa421.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "5ad0a61eabda43af4f4d7019f39e5f114a5f7b5ecaabea3682b956bc4564dbe1", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "98cb7bd5d9cdf72df673e9f8891b11c033f0a6b0714b9bb074d4ba5dff90fb87", "generation_seconds_batch": 16.82530953292735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0313587188720703, "target_z": 0.28710945716608544, "ce": 6.229123115539551, "genuineness": 4.545003414154053, "identity": 0.7246668338775635, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5ad0a61eabda43af4f4d7019f39e5f114a5f7b5ecaabea3682b956bc4564dbe1.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "5ae0b5726d2c4a67df98e276f557e643e3c462fcb1fbcdbaae55f52231bc899d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e95e5b227d7a5325076b49e796138120810d3887d052cb5fe3828deaec1465c3", "generation_seconds_batch": 23.165404631989077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9582583904266357, "target_z": 1.0156250395275601, "ce": 6.395919322967529, "genuineness": 4.15588903427124, "identity": 0.7193751931190491, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5ae0b5726d2c4a67df98e276f557e643e3c462fcb1fbcdbaae55f52231bc899d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "5b2f4ea4c4ba3cbae3686318b98eef5ac1c1334d62fc316d28bff4b6fb3fb4f3", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e8edabec0158ab7ee3072fe5cb6dec997eb0e4824cf98a95a90c4f7d0c6ec23a", "generation_seconds_batch": 18.18009413301479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.109982967376709, "genuineness": 4.168440818786621, "identity": 0.5624063611030579, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5b2f4ea4c4ba3cbae3686318b98eef5ac1c1334d62fc316d28bff4b6fb3fb4f3.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "5ba511301e1ad0cb6336d12af4e0ff4e22c330a7f54db9b4cd74a39e5300db3b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b06d7f88df97a1418c1d829a4acd68f92a9da27f3f87deb6423cb6a3b1a7fada", "generation_seconds_batch": 22.95536903105676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9965686798095703, "target_z": 0.25976552087415894, "ce": 6.1894097328186035, "genuineness": 4.557555198669434, "identity": 0.6425212621688843, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5ba511301e1ad0cb6336d12af4e0ff4e22c330a7f54db9b4cd74a39e5300db3b.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "5c8a838dc537270907a6604bfdc9662639edba9292c4179107b0beee36c0dd6e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "81b39439c10545e22684e7a54d50e4da4252f9e63a9d1f6bfc4a2a0957d23934", "generation_seconds_batch": 23.529472123016603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7544894218444824, "target_z": 0.8554686888786803, "ce": 6.316492557525635, "genuineness": 4.331617832183838, "identity": 0.6036864519119263, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5c8a838dc537270907a6604bfdc9662639edba9292c4179107b0beee36c0dd6e.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "5c8be1e0dae850fb354fdfcf1fa7eb41e0773865c3046f918aa47cefbe21ef3b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "354ce7793f6b6b87457b3376474ca26b2277cfbf781b3b06d48c9a68312d6844", "generation_seconds_batch": 22.735376843018457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.524835109710693, "target_z": 1.460937494876057, "ce": 5.414999485015869, "genuineness": 4.419482707977295, "identity": 0.71590256690979, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5c8be1e0dae850fb354fdfcf1fa7eb41e0773865c3046f918aa47cefbe21ef3b.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "5caf585bbe5ce844ecc20fb243f9dc68a1aa03e31196d70fcd37d6cc9093e3b0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "82e078d6e590faab7e7fa0f5f83eece35c9b03cc18e247be116b56e2d27c4806", "generation_seconds_batch": 23.009457394015044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994083881378174, "target_z": 0.2578125431875193, "ce": 6.268836498260498, "genuineness": 4.011539936065674, "identity": 0.19576634466648102, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5caf585bbe5ce844ecc20fb243f9dc68a1aa03e31196d70fcd37d6cc9093e3b0.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "5cca2d7d7123f067c4f50680856b39e92df0976c1eed997c2655b33d1092715d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c4705c52b8fd6224f0e9ba14a5451cf3ad0d98fff7c45e6be5013d02366982f3", "generation_seconds_batch": 16.474709944915958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.438975989818573, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5cca2d7d7123f067c4f50680856b39e92df0976c1eed997c2655b33d1092715d.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "5d036a0af3b47b1fbacdbd8a68d5647490743fdc5093e6d04e3bcc88efd42ed6", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a35237c3cd7d0f746e3c164c5d142251a773909f984e2f9bbdab73a9a952cd1e", "generation_seconds_batch": 18.86498992203269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.0305562019348145, "genuineness": 4.406930446624756, "identity": 0.7061883211135864, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5d036a0af3b47b1fbacdbd8a68d5647490743fdc5093e6d04e3bcc88efd42ed6.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "5d308534f1f6fd85037ac08651a535bd09e4552a27d227d3ae4f7e0e41a8c50b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "699136a9421560d29282871780454c82a7410f69e6596c8a7113eb2da7d1b5ad", "generation_seconds_batch": 16.420581302954815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5954504013061523, "target_z": 0.7304687474380285, "ce": 6.1020402908325195, "genuineness": 3.773050308227539, "identity": 0.21439510583877563, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5d308534f1f6fd85037ac08651a535bd09e4552a27d227d3ae4f7e0e41a8c50b.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "dae472a8d75b99630af159da19cc204730d7d6b548bdba69e7ad45593c40b7db", "generation_seconds_batch": 15.186213168082759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.341981887817383, "target_z": 0.5312500322076416, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.7397088408470154, "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_S_MONO_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "5e13820f9c61b74dd94a4f02199d5d527a57084eea8768cd61cb1089ef05a5dd", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1ce658344e0f69093407f4aa96faf4625b2e5b44f7a59382e7c557e5ff74ce64", "generation_seconds_batch": 16.832905864110216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.6618211269378662, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5e13820f9c61b74dd94a4f02199d5d527a57084eea8768cd61cb1089ef05a5dd.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "5f6022a46b214466aee623dfb5320005992dfdc1167bab899439f3b12f3a23a4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "983a77ec77a6e64e4700e773c0cd0f2eb3dc629be79462e4de8198333bd1923c", "generation_seconds_batch": 23.812236011959612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.206757068634033, "target_z": 1.2109376119947535, "ce": 6.332377910614014, "genuineness": 4.507347106933594, "identity": 0.7891638278961182, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfahren.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5f6022a46b214466aee623dfb5320005992dfdc1167bab899439f3b12f3a23a4.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "5f7c8258f15177b0f767043dd6423d235c64fbb7bd767b2680cc1b649fef76e4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0880d9f261811f977ca52fb4fb70b8dee6f7b178ac5bfa97c2d4ac0fd786b397", "generation_seconds_batch": 16.18322566500865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197852611541748, "target_z": 0.41796870644648476, "ce": 5.903473377227783, "genuineness": 4.168440818786621, "identity": 0.697107195854187, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5f7c8258f15177b0f767043dd6423d235c64fbb7bd767b2680cc1b649fef76e4.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "5fe3f1ed09008fb3233cdec209ff8e2044aa9949203bbf9dd436e87774a956cf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cdf191f74461469c316af340fed081881af3c9d8c9d5b6c134a4367fb404a123", "generation_seconds_batch": 22.495255694026127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 5.990842819213867, "genuineness": 4.406930446624756, "identity": 0.7489862442016602, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/5fe3f1ed09008fb3233cdec209ff8e2044aa9949203bbf9dd436e87774a956cf.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "6049ea9f1897ad974a4bbd61f5aa7726534fdacaa7f9ef255bb02f665fdf69ff", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b0edc2c3625259a1b2d50c6859d64d2490511c312c76ed173c2cafe40e9eaaa5", "generation_seconds_batch": 16.88931248593144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.351921796798706, "target_z": 0.5390625051239429, "ce": 6.0305562019348145, "genuineness": 4.005263805389404, "identity": 0.5886495113372803, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6049ea9f1897ad974a4bbd61f5aa7726534fdacaa7f9ef255bb02f665fdf69ff.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "6084881d8709fa9ae434fcbcb64a456e0b03784a6c56ab949e17feaa12a4a37c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "11e7b46cec7ae65f7fbd210dce99e4f20262d68589317965aeba7d8ee0350e21", "generation_seconds_batch": 17.92705337505322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3320419788360596, "target_z": 0.5234375592913401, "ce": 6.356205940246582, "genuineness": 4.337893962860107, "identity": 0.6055036783218384, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6084881d8709fa9ae434fcbcb64a456e0b03784a6c56ab949e17feaa12a4a37c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "60a85752fdd74cc7221b06024f2b666a68109379f88c1c2bb89f960dc0e1899d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c0b4e97febab8a043a1e6acb167b752820b01d0455360d7442609bf44d36a48d", "generation_seconds_batch": 23.991533047985286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 5.95112943649292, "genuineness": 4.337893962860107, "identity": 0.5655895471572876, "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_S_MONO_reg/60a85752fdd74cc7221b06024f2b666a68109379f88c1c2bb89f960dc0e1899d.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "60bd9aebf6e2772ef8fb79af64348bbe9b74102e3bfb614d1d27a5276019f19f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cb00d81e6d33ef0130676a14f7e405663498dd48f635ff123caf0a16397e5021", "generation_seconds_batch": 22.859746392001398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 5.80021858215332, "genuineness": 4.457139015197754, "identity": 0.8013577461242676, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/60bd9aebf6e2772ef8fb79af64348bbe9b74102e3bfb614d1d27a5276019f19f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "6157bf5ebb2dd42ebe31ccfc9373d6326a0c41f07a2f1cb3bad5349beb7230b5", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fbdf0307d857b809783148866546a11b0eddac0ceb2ad37f2e060da19f1422b2", "generation_seconds_batch": 17.253830702975392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9135284423828125, "target_z": 0.9804686303193321, "ce": 6.062326908111572, "genuineness": 4.595211505889893, "identity": 0.6047756671905518, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6157bf5ebb2dd42ebe31ccfc9373d6326a0c41f07a2f1cb3bad5349beb7230b5.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "61590d5997b2f3d4c33579b37714a0d25beab6966e9f96f3219c158758814ee2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e34af93407831b1860978b8919233bc817e610b36a613fb86430c022f22b5289", "generation_seconds_batch": 18.44772389996797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3022220134735107, "target_z": 0.49999995315252144, "ce": 6.260893821716309, "genuineness": 4.582659721374512, "identity": 0.5727185010910034, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/61590d5997b2f3d4c33579b37714a0d25beab6966e9f96f3219c158758814ee2.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "61a37445472613c2b8a820dbaa1854bee1e2aa20fc4a8597b372922c51bfbf0e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "21af82963df18818027216611bda88c69101a1ef44718ec0cb9391050a4ff69a", "generation_seconds_batch": 22.859500022022985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 5.871703147888184, "genuineness": 4.250029563903809, "identity": 0.5565067529678345, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/61a37445472613c2b8a820dbaa1854bee1e2aa20fc4a8597b372922c51bfbf0e.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "61ebfb1b4373d6ea868a7b915850c9d4a4140d68d1b8b764b8737b4cdd24be7d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "148e6182d08f3014dd7702fba8f86baf2def3df4080bc98b178c32d3db3dabe4", "generation_seconds_batch": 23.794503704993986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.7260102033615112, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/61ebfb1b4373d6ea868a7b915850c9d4a4140d68d1b8b764b8737b4cdd24be7d.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "620993c4a48fbb106fdfe2dabe62be67ce734287e3068132d27ba731cabef76f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c22889519796ffc54b98b286eee2814b5eb718cf27e25b8c51ffee24b7226bfa", "generation_seconds_batch": 17.39460976794362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197852611541748, "target_z": 0.41796870644648476, "ce": 6.054384231567383, "genuineness": 4.425758361816406, "identity": 0.652098536491394, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/620993c4a48fbb106fdfe2dabe62be67ce734287e3068132d27ba731cabef76f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "622a222e298478526f04541ccaa31c6ca3e7edf98789a35f98141f8322ec138e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cd67f7b8b49675c7bdeb0cc86d1d1a2b6bdba4feacb4a3e3d3916fd41ae2315f", "generation_seconds_batch": 16.357562349992804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.341981887817383, "target_z": 0.5312500322076416, "ce": 6.133810997009277, "genuineness": 4.218649387359619, "identity": 0.5923199653625488, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/622a222e298478526f04541ccaa31c6ca3e7edf98789a35f98141f8322ec138e.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "62948e234c07003427d6d5b62571438d50c69aed8b486bb885f5b72146be5d14", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c6584908390512432b2b77235a5dd6492c88eaa26fee47ffad852f4696f9df64", "generation_seconds_batch": 23.256876711035147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545750617980957, "target_z": 0.691406195466607, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.6775215864181519, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/62948e234c07003427d6d5b62571438d50c69aed8b486bb885f5b72146be5d14.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "629558da478e3c1ec41f3eaa1305d5fa11a31635d0f10c87037607ba48063be6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4d439473bb29c5593b196f765d72da4720f243a7d76c4e66050257ed501ab5b8", "generation_seconds_batch": 23.794210357940756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.107357501983643, "target_z": 1.1328125080519105, "ce": 6.41180419921875, "genuineness": 4.287685871124268, "identity": 0.5399404764175415, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/629558da478e3c1ec41f3eaa1305d5fa11a31635d0f10c87037607ba48063be6.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "6298b8fb0b96588e45a5e4c8a7617f9bc7ccc9569faeefa637230f9c1b089569", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c90bbf489c3501bc9ef8d4e456bf22663fc895cea861b55a83d305becedd824e", "generation_seconds_batch": 23.686179817072116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.968198299407959, "target_z": 1.0234375124438615, "ce": 6.054384231567383, "genuineness": 3.7291181087493896, "identity": 0.5625534057617188, "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_S_MONO_reg/6298b8fb0b96588e45a5e4c8a7617f9bc7ccc9569faeefa637230f9c1b089569.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "62d4311a43c847314ff55c9fc4724cb00d62cd9de0a3c4842be097d9dbe46d34", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "21858fe6f8305dfcbfd1a8b3450ce67dfabbce35c6c977b82fac2b5a753a309c", "generation_seconds_batch": 17.11183617694769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.006728172302246, "genuineness": 4.21237325668335, "identity": 0.5408900380134583, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/62d4311a43c847314ff55c9fc4724cb00d62cd9de0a3c4842be097d9dbe46d34.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "62dac3492b2562ec31554a31578a099de2682b635ac2be49465ffc94617cc7fa", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2fbe54f20e4f0cb91a8d5e9016667ade89db354b2986bc68506caf835f3c4def", "generation_seconds_batch": 15.092310701031238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.022613525390625, "genuineness": 4.482243061065674, "identity": 0.8822605609893799, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/62dac3492b2562ec31554a31578a099de2682b635ac2be49465ffc94617cc7fa.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "62eed21ec74adf96d31173777b3b5f2770ac1601e7c4bed1a1f60d16ed7552c7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7c3e624c25ca9a13ae0110be83aab59edbcfa2abf368c84fe70011f374b78f00", "generation_seconds_batch": 23.770009348052554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2699172496795654, "target_z": 0.47460936932706316, "ce": 6.046441555023193, "genuineness": 4.400654315948486, "identity": 0.5947357416152954, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/62eed21ec74adf96d31173777b3b5f2770ac1601e7c4bed1a1f60d16ed7552c7.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "631eb677318933ea4e367dc45f63a3431b5b229053309fd2450c07acad00ded7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "25e42c97c9e56e55a27e4c8afd745631ae171a34c799a3a24b931cdc6dc9b235", "generation_seconds_batch": 24.75043791590724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.41180419921875, "genuineness": 4.632867813110352, "identity": 0.5393966436386108, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/631eb677318933ea4e367dc45f63a3431b5b229053309fd2450c07acad00ded7.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "6323636047759b0069b07c28a4ff5d6cf48be128828fad4376f1e3a1500b45e4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2396afd38d0163547123ba6c6a7f7cd4e8b2d9b0d5ddbd3cc3166797a06b4224", "generation_seconds_batch": 25.46872396604158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 5.824047088623047, "genuineness": 4.231201171875, "identity": 0.5256506204605103, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6323636047759b0069b07c28a4ff5d6cf48be128828fad4376f1e3a1500b45e4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "63cd7451a93304774b976efd0a5d0c88a664875facbb9abd4213c610890d64e1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "090f2adf915991970a6c272a13912204e05d3b9cb4d0bb8e363833ac4956ba53", "generation_seconds_batch": 22.943207517033443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 5.998785495758057, "genuineness": 4.65797233581543, "identity": 0.7562251091003418, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/63cd7451a93304774b976efd0a5d0c88a664875facbb9abd4213c610890d64e1.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "63d3d673a6e6a1406344ac1e0409ad93f025e4ed02fdb1d72efc7da47e31250d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "526f9c03fe6f1756af6597b4e520e58b39cc2c06c7dccc0380c5a4226f0372d4", "generation_seconds_batch": 23.546955268946476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.694849967956543, "target_z": 0.8085938513808716, "ce": 6.054384231567383, "genuineness": 4.758388519287109, "identity": 0.7062604427337646, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay?", "wer": 0.058823529411764705, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/63d3d673a6e6a1406344ac1e0409ad93f025e4ed02fdb1d72efc7da47e31250d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "6422bbad7b714045b33b7835364b9adea39853b5d5f696d74b74d569c7dd7c19", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bb0f91a3fb0571ac6152d123519ce012c806c14c7f3366511276536f18d844aa", "generation_seconds_batch": 17.465819238917902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.1020402908325195, "genuineness": 4.293961524963379, "identity": 0.6743365526199341, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6422bbad7b714045b33b7835364b9adea39853b5d5f696d74b74d569c7dd7c19.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "6426a16fbcd892f8efeee873f7b4dedf0b38c4276b06c7690fed98cc6965794b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eac87bf2ac338d0b6d1a779e8acaf13afb0ebc3e15c429d72b2fcb45e548befd", "generation_seconds_batch": 17.082865569042042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964263916015625, "target_z": 0.23437493704870066, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.30484920740127563, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6426a16fbcd892f8efeee873f7b4dedf0b38c4276b06c7690fed98cc6965794b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "6488998fba07441d18937dd270f6cfc998b409bf1d2175e98249a0dd0d6c8862", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f86a4930e7da2ae984dc74edc396e3fb6e39d9806852d866e6faac630eb60f78", "generation_seconds_batch": 19.951789537095465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 6.332377910614014, "genuineness": 4.325342178344727, "identity": 0.49300435185432434, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6488998fba07441d18937dd270f6cfc998b409bf1d2175e98249a0dd0d6c8862.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "64cc09274317859a93234be96a699b2ea82ba0820180fd7be7e0d8518be752ba", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b35f9449c3b5fa18402e183c951dee2ca02763fc6b390053afb59c00e3e96484", "generation_seconds_batch": 22.771051340037957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.0305562019348145, "genuineness": 3.9801597595214844, "identity": 0.5299411416053772, "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_S_MONO_reg/64cc09274317859a93234be96a699b2ea82ba0820180fd7be7e0d8518be752ba.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "64d2119ec0fab433415a07bca3be5f531b0662d93dcac4442ccce63c4dc6e939", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9110ea1f8b5ceb27139f336fe5ca2fc76706176abfa689b4d72f77f511d01010", "generation_seconds_batch": 16.408988456008956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.779132604598999, "target_z": 0.08886723077901829, "ce": 6.09409761428833, "genuineness": 4.337893962860107, "identity": 0.5568555593490601, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/64d2119ec0fab433415a07bca3be5f531b0662d93dcac4442ccce63c4dc6e939.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "64e73339c8fcc1676a45d717be53bdd0821e7f1822932a66de7c443046135f94", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7af6a5ff253dec63d98e6e7e206588d9903c0d4bf455019b30ba0ec1bc372d58", "generation_seconds_batch": 17.693446008022875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1332430839538574, "target_z": 0.3671875387955682, "ce": 6.181467056274414, "genuineness": 4.582659721374512, "identity": 0.6650737524032593, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/64e73339c8fcc1676a45d717be53bdd0821e7f1822932a66de7c443046135f94.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "65550bf5a73525df7a2229d1f52b6c9a441cccb3719aea82a86f85b21a03c33c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "57604db397a2fbae96d5328f15407d4df727b4a2b771106281c5c8abf05e9e2c", "generation_seconds_batch": 16.968194245011546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8837087154388428, "target_z": 0.9570312115704277, "ce": 6.038498878479004, "genuineness": 4.2751336097717285, "identity": 0.7940438389778137, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/65550bf5a73525df7a2229d1f52b6c9a441cccb3719aea82a86f85b21a03c33c.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "657cb39e2a3e89b6f90511ecc8e9f0c835d8f4473f9451bebd60a847ca33ab8e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4e6fe0ecc2cb15b0bbb1000fcbc8c27acb6befc98dabf303e64f253b377ba56c", "generation_seconds_batch": 25.741844723932445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.348263263702393, "genuineness": 4.444586753845215, "identity": 0.4190645217895508, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/657cb39e2a3e89b6f90511ecc8e9f0c835d8f4473f9451bebd60a847ca33ab8e.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "65d19cbeadc36ea33a56640e4da259d092681bb49fe01d3a70ed318cad158530", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "deebb0248c19c33e501f248a52bbc2b146a9b6566a77d803f85a6addfc8b26ed", "generation_seconds_batch": 17.89656670903787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.475345611572266, "genuineness": 4.400654315948486, "identity": 0.5203108787536621, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/65d19cbeadc36ea33a56640e4da259d092681bb49fe01d3a70ed318cad158530.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "65f64527608a898ce85c74f228a628437525c2511585ea10a3f095042ec5a46c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "48abdaa4fa5a186a01feafb2e50cdf1d48de932bc09df290e6a934f4438b12f2", "generation_seconds_batch": 22.495255694026127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.793213844299316, "target_z": 1.6718752005657675, "ce": 4.851565837860107, "genuineness": 4.293961524963379, "identity": 0.5999263525009155, "hypothesis": "Juventus Boite Press.", "wer": 1.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/65f64527608a898ce85c74f228a628437525c2511585ea10a3f095042ec5a46c.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "664dd3629e43174e1dd17c3de9713bca755613c15da3ecaf6dea9d8166742300", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1271bedd54206ad8d5d2861977ff57d2c231c1f052a09de8995ee1878902f39d", "generation_seconds_batch": 17.21253404498566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.773333549499512, "target_z": 1.656249879953336, "ce": 5.824047088623047, "genuineness": 4.7834930419921875, "identity": 0.7847081422805786, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/664dd3629e43174e1dd17c3de9713bca755613c15da3ecaf6dea9d8166742300.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "665e1d9e111a13fce4fdb585f79e30282cf72fd62106d72897489e2b7944407f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "75f39fa012a6d7474e4f9a5e096e1ae4e438236dc09a1d740a3ddc6d24d3ed3c", "generation_seconds_batch": 17.101057136896998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2962164878845215, "target_z": 1.2812500556313808, "ce": 6.0305562019348145, "genuineness": 4.582659721374512, "identity": 0.7773182988166809, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/665e1d9e111a13fce4fdb585f79e30282cf72fd62106d72897489e2b7944407f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "66dd5c008d926f293231fc9f3f93dbc9033584fc337fa4581db67fb69cb655f8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c729a0216ce68796b4c036143e5acf247e7590c1995d9a70464486400fde7ade", "generation_seconds_batch": 23.920740696950816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 5.927301406860352, "genuineness": 4.218649387359619, "identity": 0.5780196189880371, "hypothesis": "De deur 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_S_MONO_reg/66dd5c008d926f293231fc9f3f93dbc9033584fc337fa4581db67fb69cb655f8.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "67a0058348ff2dd32d3af13036d15004c9b59f53af181a7e7b8477fa86056168", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6c3b7b5ae0d8c105391eefee659448edeef1a728c7cae4764ebfb71ce97dc1a9", "generation_seconds_batch": 20.549166548997164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.157639026641846, "genuineness": 4.846253395080566, "identity": 0.5567122101783752, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/67a0058348ff2dd32d3af13036d15004c9b59f53af181a7e7b8477fa86056168.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "67d3c7be2190865240771b5af41d4dd34a0f42b33b05c95f479295ba0ca27f88", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b801eb28796f305d9019c7c04bda6a0c1b1a3e5d85a662ad88f57cf2c9b17bb4", "generation_seconds_batch": 16.988274784060195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6749699115753174, "target_z": 0.7929687181583545, "ce": 5.704906940460205, "genuineness": 4.3881025314331055, "identity": 0.6569392681121826, "hypothesis": "Die Tür 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_S_MONO_reg/67d3c7be2190865240771b5af41d4dd34a0f42b33b05c95f479295ba0ca27f88.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "68072100da31e2acd1adb6b5755594bee130f68da02f27c5d3049088f0d6ba9d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "87bcc397791fead710412a6f412ae7bb76d55e3621093fa88dc24bbf41843242", "generation_seconds_batch": 22.215440520085394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.06759786605835, "target_z": 1.1015626163867045, "ce": 6.197352409362793, "genuineness": 4.7458367347717285, "identity": 0.6758140921592712, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/68072100da31e2acd1adb6b5755594bee130f68da02f27c5d3049088f0d6ba9d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "68b1f0ffb3fe9747f1b2fcd11a61979c944fe3966a541e90b4c95318679209ad", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d7f2a5baf02b43469a66033910e4f8f6dd7cd8baa49ec6922f6f6245832c05c5", "generation_seconds_batch": 25.010454149916768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 5.911416053771973, "genuineness": 4.432034492492676, "identity": 0.6183649897575378, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/68b1f0ffb3fe9747f1b2fcd11a61979c944fe3966a541e90b4c95318679209ad.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "68b6d876e4d6f61541be78b6d6a00069bca4dc441b672ebea9632b756e5fb24d", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2ba088ac5492c1a7090c419d89174c0429b06393aeaa66b9cb2cb27641e341b", "generation_seconds_batch": 27.431504456093535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.594414710998535, "target_z": 1.5156249926800816, "ce": 6.1020402908325195, "genuineness": 4.482243061065674, "identity": 0.6661276817321777, "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_S_MONO_reg/68b6d876e4d6f61541be78b6d6a00069bca4dc441b672ebea9632b756e5fb24d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "68ce45825b8fd8d493235880b1b774ec54ac73ebdf35c5e32ca7b4434a832369", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "32f8c06596cde84ae125e40c042629c26bfba9b59ae466dbd75f0745e93305fe", "generation_seconds_batch": 24.163314861012623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 5.903473377227783, "genuineness": 4.325342178344727, "identity": 0.6907479763031006, "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_S_MONO_reg/68ce45825b8fd8d493235880b1b774ec54ac73ebdf35c5e32ca7b4434a832369.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "6932a294abc0dde61c9a220463b81289cf0d3ee438ec658fba610bb327c34f12", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e74c17052b1f50753dd288a91f1492d8490877e22ea647e4570c23da052b6faa", "generation_seconds_batch": 16.946800643927418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4364113807678223, "target_z": 0.6054688059973767, "ce": 6.3641486167907715, "genuineness": 4.394378662109375, "identity": 0.3954031467437744, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6932a294abc0dde61c9a220463b81289cf0d3ee438ec658fba610bb327c34f12.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "69be4479cfbaa326feb629105a54280f283d50209fc6602d3586bffe25d4a4fe", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba03d578e2057a5f6468df37a9fc2e02d935dfbe3b98b827bcb5c972836e812d", "generation_seconds_batch": 16.90731485199649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435375690460205, "target_z": 1.3906250512394296, "ce": 5.839932441711426, "genuineness": 4.457139015197754, "identity": 0.7882179021835327, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/69be4479cfbaa326feb629105a54280f283d50209fc6602d3586bffe25d4a4fe.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "6a4166425413a576f80955cda281016b8761d62c65ab3bbe029e0689d9315b47", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "01aca882375a2d6b5eeed7a738e30c5dc2d6d5f2ef26168f42a9ac20685934c4", "generation_seconds_batch": 15.614239769987762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2177324295043945, "target_z": 0.4335936522790877, "ce": 6.348263263702393, "genuineness": 4.061748504638672, "identity": 0.20509153604507446, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6a4166425413a576f80955cda281016b8761d62c65ab3bbe029e0689d9315b47.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "e48100451d9b1ee6d2da159ce89b256ccb39f637c7350817ececcaa08953d42f", "generation_seconds_batch": 15.105511959991418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0537235736846924, "target_z": 0.3046875680752423, "ce": 6.3641486167907715, "genuineness": 3.898571252822876, "identity": 0.2320227324962616, "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_S_MONO_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "6a8e87325ace3c8ef50067261a4d07b2292caf0128f21cd1e9fe97a621ee9f61", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "617dde055213bf543d09444227cc261be48473f253fe510e12a8c5eec455b2d0", "generation_seconds_batch": 16.101050292956643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 6.23706579208374, "genuineness": 4.770940780639648, "identity": 0.5727143287658691, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6a8e87325ace3c8ef50067261a4d07b2292caf0128f21cd1e9fe97a621ee9f61.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "6a91c2e24c4afbe31edd5f02055e2c08ee44b6ffdce05418d67a08b94aa753be", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe8b3c3f4f9d1a4c84684c32515a2da9f15f1c2433861cc196455894f8f38b04", "generation_seconds_batch": 27.782143476069905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3718016147613525, "target_z": 0.5546874509565459, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.6138283014297485, "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_S_MONO_reg/6a91c2e24c4afbe31edd5f02055e2c08ee44b6ffdce05418d67a08b94aa753be.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "6aaf6bb111499d58e3f683d9c4843e01382eafc776a9f4fc2aaa9de5bb222625", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b1a02ef27a1d559df4679f9d33604e477434979fa2ed01e312ead230e6b8e101", "generation_seconds_batch": 15.641998409992084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.982072830200195, "target_z": 1.8203127481452381, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.6571789383888245, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6aaf6bb111499d58e3f683d9c4843e01382eafc776a9f4fc2aaa9de5bb222625.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "6ad296897fa93f9c436f6e0b105c30d22fa49cc46cc90a9e3d9cc26c9c82c392", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4a83284bb02f0ce7952a071d0261415f6e283a17ef317aebe5b684ad35dee3af", "generation_seconds_batch": 23.702739716041833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.4602317214012146, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6ad296897fa93f9c436f6e0b105c30d22fa49cc46cc90a9e3d9cc26c9c82c392.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "6b467d60e9c3d2e7fe449181c15e46f1313fdc233e374faa3fcceb4f4516e3f4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "814573319f6ecde015f670233daf17ddb32d2f12f06956f1d030688d8e0edd7b", "generation_seconds_batch": 22.675014531007037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4761710166931152, "target_z": 0.6367186976625825, "ce": 6.141753673553467, "genuineness": 4.375550270080566, "identity": 0.6859409213066101, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6b467d60e9c3d2e7fe449181c15e46f1313fdc233e374faa3fcceb4f4516e3f4.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "fc88b272067ec24ca4b1114a99c251349cb9f845eacc159c7d88ffc71e4c4ba9", "generation_seconds_batch": 16.697768932092004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06117844581604, "target_z": 0.3105468759149898, "ce": 6.149696350097656, "genuineness": 4.6956281661987305, "identity": 0.7025009393692017, "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_S_MONO_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "6bb06735c01cfc1a4e0e2d6598bfad7f3a997876dc260035a627fbfc38e7c59a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b385fee3a19bb74b3ea498c42d60eeb869d9bab42ef63dd002a2bd4f31cd39a0", "generation_seconds_batch": 26.52677231898997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.0305562019348145, "genuineness": 4.482243061065674, "identity": 0.4975856840610504, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6bb06735c01cfc1a4e0e2d6598bfad7f3a997876dc260035a627fbfc38e7c59a.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "6bb0ee8343099010e786711308d2cb38268320f9b8a5c80340f112d45a9f5448", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb8432d12dcba11ed69b120899bc37be7149a1c303fb868fbe2042a6c64a2673", "generation_seconds_batch": 20.45865741802845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.791414737701416, "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_S_MONO_reg/6bb0ee8343099010e786711308d2cb38268320f9b8a5c80340f112d45a9f5448.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "6c517722b38d59008ac901175561bd39827eff022a8948fc1a68722dd8117e77", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "eb03fb93bc3f061fe4f41bb59090fa5a88b7d77ebea217b178b886178da75658", "generation_seconds_batch": 20.657919200952165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 6.300607204437256, "genuineness": 4.545003414154053, "identity": 0.5140218138694763, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6c517722b38d59008ac901175561bd39827eff022a8948fc1a68722dd8117e77.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "6c5940d8542f647e758f6a348485825557873481c7203b81052f8560b3cf2a82", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8d1a56aa86e8cc6d928496f97dce4317ccbf5a36c60a5cb3e1122e81ad6309f5", "generation_seconds_batch": 23.13551482197363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6849098205566406, "target_z": 0.8007811910746558, "ce": 6.356205940246582, "genuineness": 3.9927117824554443, "identity": 0.6297853589057922, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6c5940d8542f647e758f6a348485825557873481c7203b81052f8560b3cf2a82.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "6d5cf541b67ca049b502e9300e50dd8b74b8dbe23032c0ede1002df78282fcb8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "51364d02d6762e2e43432ca22a8338841227e2081c171dfc4f2b6bee49660f60", "generation_seconds_batch": 25.246125899022445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7296395301818848, "target_z": 0.8359374128929695, "ce": 6.38003396987915, "genuineness": 4.6705241203308105, "identity": 0.2998904585838318, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6d5cf541b67ca049b502e9300e50dd8b74b8dbe23032c0ede1002df78282fcb8.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "6df031e2abc0c975a5942aa1d36b95d08547338d5fbc6cb5b92188713f98c61c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7e311366e8733cdaa43bebb789f65e73d56c467093e2f354318761fd5a449636", "generation_seconds_batch": 27.372245289036073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.6405131220817566, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6df031e2abc0c975a5942aa1d36b95d08547338d5fbc6cb5b92188713f98c61c.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "6e497da26b7cd2d9e5c4c6a15dc2159818c0689f6ce2b964b64e55402c22adbe", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6fcdaf5f7072f421b50a0528746ad721cf5a762cfcbe02fb587169fcea4fcc93", "generation_seconds_batch": 16.875278149032965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3455026149749756, "target_z": -0.2519530479578575, "ce": 6.149696350097656, "genuineness": 3.9676074981689453, "identity": 0.04005817323923111, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6e497da26b7cd2d9e5c4c6a15dc2159818c0689f6ce2b964b64e55402c22adbe.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "6ead0123410e00fb3a6f478bc2ed72c6d26eeb2f5ecb1b29f1adaf9ebf2f1cc7", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3557819b46eb3fa3983cb6ef5aed13baa70c8546adabdf0a1cfd115b77e7b9b5", "generation_seconds_batch": 21.9684867199976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9791738986968994, "target_z": 0.24609374011811, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.4983445107936859, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6ead0123410e00fb3a6f478bc2ed72c6d26eeb2f5ecb1b29f1adaf9ebf2f1cc7.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "6f8cdbde684e60a26550a87771daf5ed4e366f8a79c213d9c3208e847d2699a5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b81b057be082d80714c4f3fe6386ba5427417cf7781f72a04e0c4ae7c9ac824f", "generation_seconds_batch": 24.215460938983597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.998018264770508, "target_z": 1.0468751185826801, "ce": 6.38003396987915, "genuineness": 4.337893962860107, "identity": 0.3042822778224945, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6f8cdbde684e60a26550a87771daf5ed4e366f8a79c213d9c3208e847d2699a5.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "6f9433f85a919dc2dbe89fc60bdc7bcbdff803ea717d7517a4cfe608ddd1972d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2b7c7be7aeacb0b89a570e68f01afd7414e75d3ae7a4803dea9af0381b194f0d", "generation_seconds_batch": 22.95536903105676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.832559823989868, "target_z": 0.13085943667031358, "ce": 6.308549880981445, "genuineness": 4.507347106933594, "identity": 0.6932558417320251, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6f9433f85a919dc2dbe89fc60bdc7bcbdff803ea717d7517a4cfe608ddd1972d.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "6f9ec1c5948b3d31ee4d8e7e2be67b380264cb7db7100fc109d751a312cea4ec", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 130, "seconds_generated": 10.4, "mp3_sha256": "b71414c7e80ff3d61a1106bb8d1a4c53f14cf74daca7152c0e37654000c68313", "generation_seconds_batch": 22.293837398057804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.266396522521973, "target_z": 1.2578124494925622, "ce": 5.6254801750183105, "genuineness": 4.65797233581543, "identity": 0.592696487903595, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "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_S_MONO_reg/6f9ec1c5948b3d31ee4d8e7e2be67b380264cb7db7100fc109d751a312cea4ec.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "6fb20fb628350b2d57bfc09d82e117de6ed294d2bf21b1625ae1e7b4b5c18b9a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "380e63f97c8309f684441f141f3f47b535f9a88cbb6250d6e3bc92d022c90566", "generation_seconds_batch": 17.699867147020996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 5.752562999725342, "genuineness": 4.049196243286133, "identity": 0.6469746232032776, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6fb20fb628350b2d57bfc09d82e117de6ed294d2bf21b1625ae1e7b4b5c18b9a.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "6feae7e0f88bd2db6fb3bee1f28933698ec4069adf1b25482d80682910118be9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "de334a43f2bf4d292103be3e61bf5d6766121f17ddb089147d09bed8f45491db", "generation_seconds_batch": 17.848345541977324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.165581703186035, "genuineness": 4.369274139404297, "identity": 0.7980044484138489, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/6feae7e0f88bd2db6fb3bee1f28933698ec4069adf1b25482d80682910118be9.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "7044e43c9d9caf6ca7489be1a2dda934f737b761ef0015eca695063cbc1f877d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d422eca3826dff25eb1a887532451d86b7468af3d6ccdc8dbffd78cc68e6a27d", "generation_seconds_batch": 23.08076905296184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.395919322967529, "genuineness": 4.381826400756836, "identity": 0.2160329967737198, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7044e43c9d9caf6ca7489be1a2dda934f737b761ef0015eca695063cbc1f877d.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "70ee3eea6dca0743646e1c2ba0a6eae06fbe49e49aa178ca747d89a4e89e6eac", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6a7397dc9a50244a8a4eac4a45ff5d2e03748b14962f8e728997a9b79e4013ff", "generation_seconds_batch": 16.977997447014786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 6.419747352600098, "genuineness": 4.080576419830322, "identity": 0.5654129385948181, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/70ee3eea6dca0743646e1c2ba0a6eae06fbe49e49aa178ca747d89a4e89e6eac.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "7169627d7f2aa2540414cce7061abd2a3572f4fb56d959112ac2823895f2e832", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1375cb73e1f41c9cc416c0d09947d2da7054b318e1ab5d17544ac3fb280741d", "generation_seconds_batch": 27.68021088000387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.594414710998535, "target_z": 1.5156249926800816, "ce": 6.070269584655762, "genuineness": 4.469690799713135, "identity": 0.805058479309082, "hypothesis": "Ton message m'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_S_MONO_reg/7169627d7f2aa2540414cce7061abd2a3572f4fb56d959112ac2823895f2e832.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "71d00b958c3ba104df8cc4359c727918d7475f600615168db8aa5c404300f6f0", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81f1c7e12c670997d4af8eb3e22aa967ea19d262af48ab2cd6c9a0c5a05a58f8", "generation_seconds_batch": 27.785415231017396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9582583904266357, "target_z": 1.0156250395275601, "ce": 6.070269584655762, "genuineness": 4.394378662109375, "identity": 0.8324383497238159, "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_S_MONO_reg/71d00b958c3ba104df8cc4359c727918d7475f600615168db8aa5c404300f6f0.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "71e3b5686e228502fc35dbc3d52155b63344dfd50243b790ba1b9241d8e7718a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "56fa227164a072941c1c959af815d43f3ecf396537f1ea42e8965a96e4b7a9ab", "generation_seconds_batch": 15.95428179006558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6799397468566895, "target_z": 0.796874860921548, "ce": 6.459460258483887, "genuineness": 4.143336772918701, "identity": 0.5615019798278809, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/71e3b5686e228502fc35dbc3d52155b63344dfd50243b790ba1b9241d8e7718a.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "721d2c6625aad239303bde90075592e79f552585ad67a41d4a49955e1b95268d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "962d3a253f26f3a736f657e4c74336c4ba1086d43f337dcbfdbddfde6e03db08", "generation_seconds_batch": 22.49290754098911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8340091705322266, "target_z": 0.9179688469889205, "ce": 5.911416053771973, "genuineness": 4.344170093536377, "identity": 0.6018103957176208, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/721d2c6625aad239303bde90075592e79f552585ad67a41d4a49955e1b95268d.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "728326b686c12baec95cbffc1be6ca0f5efc772d6b0ac42793313550fb25019b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "524db0bce114ad6e816d8db9d4c18ebd6726f289718191c7734754d78f5d1e73", "generation_seconds_batch": 18.12347157509066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.1894097328186035, "genuineness": 4.545003414154053, "identity": 0.7352530360221863, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/728326b686c12baec95cbffc1be6ca0f5efc772d6b0ac42793313550fb25019b.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "7297681775b797f2a9dec0e460bb9b1b72fec9362b70d520664d8579152afab5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c835024290e3c6a3b17397b404f7fafcaacedcb96a73eef545e0ffdf82ebd27f", "generation_seconds_batch": 25.102356048999354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.6405490636825562, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7297681775b797f2a9dec0e460bb9b1b72fec9362b70d520664d8579152afab5.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "72ba85ec9c89987a020ce65e93060e8be98d3c626e3c3440b70703936b7b03d4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c26b5d92e7477fe90ab95e8f222095175294c7b737e5cf0e4751410180a7aed6", "generation_seconds_batch": 20.313041946967132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.655090093612671, "target_z": 0.7773437723257515, "ce": 6.1020402908325195, "genuineness": 4.519899368286133, "identity": 0.7915199995040894, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/72ba85ec9c89987a020ce65e93060e8be98d3c626e3c3440b70703936b7b03d4.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "72dfcfa6e034afa501d74422976fd6347f0681b5e413dd8107dbcf226f498f0f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4a8bf89ea0d09b50b93cbdeb8fb97822e344362fa6129fb99e5fdc424232cb69", "generation_seconds_batch": 20.275241427938454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.848918914794922, "target_z": 0.9296874626684155, "ce": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.7621277570724487, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/72dfcfa6e034afa501d74422976fd6347f0681b5e413dd8107dbcf226f498f0f.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "7326bafca85519e33dff01271eaa8da46ec4d408651e7a46c5c49c44725b0918", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2c66b04573ec3d47a6321b8063f7f65c8dadfee8fbb7d158da9bb09249b1f192", "generation_seconds_batch": 24.473477113060653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.255007266998291, "target_z": 0.46289056625765385, "ce": 6.252951145172119, "genuineness": 4.086852550506592, "identity": 0.3944154381752014, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7326bafca85519e33dff01271eaa8da46ec4d408651e7a46c5c49c44725b0918.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "733cc80b4a27a1cd0355ce67c24d81ae0c417ee59ea006c527bd6ce4b7b9f407", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a3d4429513fd0fc6cc9f34327d21feba053005e608142456ee096c1b5675e865", "generation_seconds_batch": 24.81252883700654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.1020402908325195, "genuineness": 4.268857479095459, "identity": 0.5514518618583679, "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_S_MONO_reg/733cc80b4a27a1cd0355ce67c24d81ae0c417ee59ea006c527bd6ce4b7b9f407.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "735f07aef1b6479004bb74d1eb87fbfc67a4e6ce98dd78e31081de2d162de8f3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "01506361c89df4e9db464f58a8518c4c503936c4bab590f9e862cf88cdda9086", "generation_seconds_batch": 24.221133627113886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 5.7684478759765625, "genuineness": 4.287685871124268, "identity": 0.6083930134773254, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/735f07aef1b6479004bb74d1eb87fbfc67a4e6ce98dd78e31081de2d162de8f3.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "737eb072a496c208aa535229e36d006266e396313761fc3eee2519587853109d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5bb12f0862e3160f07462ebee94bd200abf56935cdaed13d382d6771cc9e80b2", "generation_seconds_batch": 24.00872556993272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.268836498260498, "genuineness": 4.319066047668457, "identity": 0.20975682139396667, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "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_S_MONO_reg/737eb072a496c208aa535229e36d006266e396313761fc3eee2519587853109d.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "73a3222e7e90de8c18e88cd4f8a0dca686e86ac81b16c04fd8a492c97e7802b1", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c251cd54fb1e10d50fb809bddf83c311c838b0a63076dac275bcab8341db96ae", "generation_seconds_batch": 16.744006821070798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774369239807129, "target_z": 0.8710936347112832, "ce": 6.300607204437256, "genuineness": 4.18099308013916, "identity": 0.3274191617965698, "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_S_MONO_reg/73a3222e7e90de8c18e88cd4f8a0dca686e86ac81b16c04fd8a492c97e7802b1.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "74bc6470efb6a0f0c1201ecb63b4320d62f0faabc2c94f49c0e738b4da72b4cb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5d625e53a84a0ea2bb4f1d1af2f70faa72deb063633f7371ab4ddd96dc4265a0", "generation_seconds_batch": 25.03464772098232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.133810997009277, "genuineness": 3.6412534713745117, "identity": 0.20216697454452515, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/74bc6470efb6a0f0c1201ecb63b4320d62f0faabc2c94f49c0e738b4da72b4cb.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "74f61f23566de078a6202b8afab2c2f09a113f15e5d5596a9969136c12e6dbc7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 130, "seconds_generated": 10.4, "mp3_sha256": "4a6ebac3b65f1f76e1d452e703e7b997ef9ca817792ed85bcf8800609ad17ed9", "generation_seconds_batch": 23.840342373005114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.952252388000488, "target_z": 1.7968747672265908, "ce": 5.49045467376709, "genuineness": 4.2374773025512695, "identity": 0.6185839772224426, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/74f61f23566de078a6202b8afab2c2f09a113f15e5d5596a9969136c12e6dbc7.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "752c1d99a88ddaaec784e19848b9d3b2660c8fb854f536af10e0a86a3fd328a0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ae8bcb439f3ab069e89318b9f08dbc3dc52736dfe9d5a62be0b0bf88ce419c24", "generation_seconds_batch": 20.23984965996351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 5.935244083404541, "genuineness": 4.7332844734191895, "identity": 0.7131997346878052, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/752c1d99a88ddaaec784e19848b9d3b2660c8fb854f536af10e0a86a3fd328a0.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "75411bebb6d13e3fa7b92b043f708f3b9ca93c9dfeb42eab359d004c38f2fd9b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7f8d95210c39b204a39bd0c60e5fc8272e3d7d113c24ab8388e833b844e75e0a", "generation_seconds_batch": 17.788812594022602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.268836498260498, "genuineness": 4.362998008728027, "identity": 0.4747297465801239, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/75411bebb6d13e3fa7b92b043f708f3b9ca93c9dfeb42eab359d004c38f2fd9b.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "755ff316edf7218641c1b1ce5196d5ee05d35d55fc6fe5783a2211893708d4b7", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "535d1e3cc29f48e4afc82d1c64d12561ddfb94c4adb7d92d61d10038fdfd35f8", "generation_seconds_batch": 15.69916506903246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.004023790359497, "target_z": 0.2656250161038208, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.7451949715614319, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/755ff316edf7218641c1b1ce5196d5ee05d35d55fc6fe5783a2211893708d4b7.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "760d741d0c3282a6ee9134d237aaa2f40b4ed9e9cc584e7aac16d3c0d11d3055", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "74c3ede7649e73a624419cbe984b28c8119fdccc93f5c714c2360a604dedfd23", "generation_seconds_batch": 15.606256076018326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.341981887817383, "target_z": 0.5312500322076416, "ce": 6.38003396987915, "genuineness": 4.143336772918701, "identity": 0.7147026062011719, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/760d741d0c3282a6ee9134d237aaa2f40b4ed9e9cc584e7aac16d3c0d11d3055.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "7629bf92cb617fdc36ba3c777f8a9a4fd94d269098eb830c2767d436ccec5bfc", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cde2a56670751bbf3b03dc68ade820808ce36ae36bd8db906306f96b0619f294", "generation_seconds_batch": 17.052611658000387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.5609176754951477, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7629bf92cb617fdc36ba3c777f8a9a4fd94d269098eb830c2767d436ccec5bfc.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "7683f8ca874bfb15df47ad6f9ff29379df3813a748840a6b75ea011e8f9c5a5b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b0489cf4ddf89a2e34df64084df2d1324551c528d8618f355e2fa08d5c73cd95", "generation_seconds_batch": 22.92863033595495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.6043548583984375, "target_z": 1.5234376529862972, "ce": 5.895530700683594, "genuineness": 4.582659721374512, "identity": 0.7198750972747803, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7683f8ca874bfb15df47ad6f9ff29379df3813a748840a6b75ea011e8f9c5a5b.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "769f2e74f4c1c85b749db5565ce5cf6733dc5c75b86ee9413e0ed60e36731893", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e30696a8faf03f6e1e973760ed0b7d9aa36fca0e522f8787aa7ef72887e369d7", "generation_seconds_batch": 23.073613230953924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.5348129272460938, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/769f2e74f4c1c85b749db5565ce5cf6733dc5c75b86ee9413e0ed60e36731893.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "76f5eea4b9e38ae95d200b849772cc4fe2cecb4caec65f12f990c6e22bf38ddd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4ff5822a6e014ffd2a03e9153c32cb4a3cd3ec0d0a967ce72ee5d1e0bc0435d1", "generation_seconds_batch": 21.089074335992336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 5.6413655281066895, "genuineness": 4.570107460021973, "identity": 0.4613823890686035, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay después.", "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_S_MONO_reg/76f5eea4b9e38ae95d200b849772cc4fe2cecb4caec65f12f990c6e22bf38ddd.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "e0810ff1523489926c5bca58a2267d4d55000756caf84c8b6ee2656bd47e0608", "generation_seconds_batch": 15.281586281955242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.173524379730225, "genuineness": 4.607763767242432, "identity": 0.7510966062545776, "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_S_MONO_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "277d44aa91b0e6b62ec7ef1038082ecadfaffc4cc2061ca82220c97250d3ddd0", "generation_seconds_batch": 16.44418358593248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.6172741055488586, "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_S_MONO_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "798b4873b519e2b1b20f6014224a28d16482de7b72fbb19b21592448d39f3340", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "55825e1f63fc298d52071da37df869e0096ff655cc4bf69351330d223e51791f", "generation_seconds_batch": 22.801500339992344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.125868320465088, "genuineness": 4.331617832183838, "identity": 0.3316854238510132, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/798b4873b519e2b1b20f6014224a28d16482de7b72fbb19b21592448d39f3340.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "79bd2e037ab3113dd94b4411e203dea68fac64382401a906398a7f5f0a50e34f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1aeeffbdcfb5fcba9768a4bec5e894b9034e27b428b5d54bde32e8aed0f51e07", "generation_seconds_batch": 22.702436299063265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.038498878479004, "genuineness": 4.256305694580078, "identity": 0.6076095104217529, "hypothesis": "De deur 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_S_MONO_reg/79bd2e037ab3113dd94b4411e203dea68fac64382401a906398a7f5f0a50e34f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "79f2eb4579e8a74d6f90780da49422cf135fb81b07baf5daac90ba54fd9d3489", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fcbab821e92da8db29d00c90ff856d0c20db5a4e24bde1e8ca49d71cea5226d0", "generation_seconds_batch": 16.88917155202944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1108782291412354, "target_z": 0.34960942788641136, "ce": 6.1020402908325195, "genuineness": 4.1998209953308105, "identity": 0.5720350742340088, "hypothesis": "De deur 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_S_MONO_reg/79f2eb4579e8a74d6f90780da49422cf135fb81b07baf5daac90ba54fd9d3489.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "7ab6694362fa378218c77304acff1d6f382fc3ada35f965b90964023598d8d0d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8f2b95dc31d94db8e4329da603af3aee09fef123c971b1e842f9aaa2f1aa1ff4", "generation_seconds_batch": 16.62805140390992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.125868320465088, "genuineness": 4.406930446624756, "identity": 0.7680454254150391, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7ab6694362fa378218c77304acff1d6f382fc3ada35f965b90964023598d8d0d.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "7ad333b840e4ad5e4147adc4d367229ca9ee9d860cded77787f66ce315d384bd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d3cc97d91ffc2d700e67b1d2870c6f1e22874422088f10d196087fa18f32bef2", "generation_seconds_batch": 22.801500339992344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.236577033996582, "target_z": 1.234375218133572, "ce": 6.292664527893066, "genuineness": 4.15588903427124, "identity": 0.31938472390174866, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7ad333b840e4ad5e4147adc4d367229ca9ee9d860cded77787f66ce315d384bd.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "7b3f4a5be375aa7c282f5d9aeeac93bc562f32d285fe089c197c76a4105668ec", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ac77317f95d4217ae77a8c322f239b571032687765ab02c44d03807b855da38a", "generation_seconds_batch": 22.91530083399266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 6.260893821716309, "genuineness": 4.143336772918701, "identity": 0.23955519497394562, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7b3f4a5be375aa7c282f5d9aeeac93bc562f32d285fe089c197c76a4105668ec.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "7b98f4b0c3d182758d034e6c2664b8de5441b9592b54b98ae059ade77995bd55", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a31f3a34ff5564ec6322f6cff1b14766b1315b199d6cb8489cd858f7d2e35bac", "generation_seconds_batch": 23.794503704993986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5656304359436035, "target_z": 0.7070311412992099, "ce": 6.054384231567383, "genuineness": 4.3881025314331055, "identity": 0.7340072393417358, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7b98f4b0c3d182758d034e6c2664b8de5441b9592b54b98ae059ade77995bd55.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "7bc3f1490c532fcdd3fb1d67f3cbc2f1a1323d45ee81080aabe956bdc0d9c41d", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "35bf6da5dcf0fb6116d9f860e9191ceb42f2771c611f02f271ca20a011982f2b", "generation_seconds_batch": 16.79256567498669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.229123115539551, "genuineness": 4.306513786315918, "identity": 0.636870265007019, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7bc3f1490c532fcdd3fb1d67f3cbc2f1a1323d45ee81080aabe956bdc0d9c41d.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "7bdf7f695825a126d8949cd286557c7a530caaf9ad5fde44ce4e32bddc908920", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "57baeeeea13adb052090f3005519f79b9bbd4e41202d95a8825fecad8cb2a89f", "generation_seconds_batch": 26.98538244701922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.427689552307129, "genuineness": 4.262581825256348, "identity": 0.49837231636047363, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7bdf7f695825a126d8949cd286557c7a530caaf9ad5fde44ce4e32bddc908920.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "7c31fc4cdd5e63ba28b9679d0158350a1da160a140abc1db53850d92cf1d57a5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7fd615a699738a8ea1aa879737bc6eb050e9ebee0bd19122efdaa7d9a73f7a98", "generation_seconds_batch": 20.181037642993033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.236577033996582, "target_z": 1.234375218133572, "ce": 6.109982967376709, "genuineness": 4.281409740447998, "identity": 0.4086270332336426, "hypothesis": "De deur 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_S_MONO_reg/7c31fc4cdd5e63ba28b9679d0158350a1da160a140abc1db53850d92cf1d57a5.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "7c67b8a4483424675d33c068e68ecfd4424d7c49e8f1986f998a634aee516808", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bf98b684a34932d70c332ee1ab9ff040dbf1f6b6e5013d16731ae996cfafd938", "generation_seconds_batch": 23.2541243910091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 5.728734970092773, "genuineness": 4.193545341491699, "identity": 0.5987870693206787, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7c67b8a4483424675d33c068e68ecfd4424d7c49e8f1986f998a634aee516808.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "7cc81e2c57729ce39e9a83ed6bf4079804501ebe631eed5a6b54b5ebf79e6b0f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2e4fdfd488e8eac89cea0c2f75c24642843f73448701cb244a5bf27224450025", "generation_seconds_batch": 28.649957439978607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.425435543060303, "target_z": 1.382812390933214, "ce": 6.117925643920898, "genuineness": 4.7332844734191895, "identity": 0.879082202911377, "hypothesis": "La porte 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_S_MONO_reg/7cc81e2c57729ce39e9a83ed6bf4079804501ebe631eed5a6b54b5ebf79e6b0f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "7cf4087c6551b1dd12db553cc7361425f84e83424fd436c5bcb71171d3576a25", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f38d172eeb4f2b42bcaea46a4fb3e5d87a8f776211ab0e2691e20e91af6522b0", "generation_seconds_batch": 17.275074098957703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 6.38003396987915, "genuineness": 4.582659721374512, "identity": 0.8767849206924438, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7cf4087c6551b1dd12db553cc7361425f84e83424fd436c5bcb71171d3576a25.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "7dc4ca66bdb69573e0abe5e9f45504a2467b79234372ca81ec5350f86228101a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a6ae432a7a171b7065a284debd3eb5937e63fcea530eb9235bbeb8a6b977a8b8", "generation_seconds_batch": 24.103951104916632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9135284423828125, "target_z": 0.9804686303193321, "ce": 6.300607204437256, "genuineness": 4.118232727050781, "identity": 0.34481459856033325, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7dc4ca66bdb69573e0abe5e9f45504a2467b79234372ca81ec5350f86228101a.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "7dcf5138bb899d4e3bbb617c97663c5b523c37dfdfc34943080abf4976500b97", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b6b20632b9cb5bfcd6e23de4ed0626df21bad2bc08b4ac224dd54e86a124732e", "generation_seconds_batch": 19.548522931989282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.4912309646606445, "genuineness": 4.356721878051758, "identity": 0.6614724397659302, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7dcf5138bb899d4e3bbb617c97663c5b523c37dfdfc34943080abf4976500b97.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "7ddb43b0bde6005836a1c36c2e1a4a32a414328163457d95cabee55fdef78439", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ba530e54c70cf0e0cb4e023aa8ad30fb0c70a5f9988bda6990f304bd7af3d1e4", "generation_seconds_batch": 22.671564646996558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986628770828247, "target_z": 0.2519530479578575, "ce": 5.927301406860352, "genuineness": 4.256305694580078, "identity": 0.6678066849708557, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7ddb43b0bde6005836a1c36c2e1a4a32a414328163457d95cabee55fdef78439.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "7e52090c2659bfe21ea2e8f48a24ca3aa7a644c0c27f44d51dcb9432deb8093e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "620642b714bf62b2c17cf20277fe8eed2ab86ac50468c304af7f50e72bb15d65", "generation_seconds_batch": 23.524299837998115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9184985160827637, "target_z": 0.98437496047244, "ce": 6.41180419921875, "genuineness": 4.6203155517578125, "identity": 0.6201485395431519, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7e52090c2659bfe21ea2e8f48a24ca3aa7a644c0c27f44d51dcb9432deb8093e.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "7e6df2cafbac04cbf4e3627e5bdfa3b7f179d61889a2ab14d2b2e4ec91e484fa", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0ccb14445bda15ded54db512c8182ac2e53410c9eb02db655bf0bbb043fe8bbb", "generation_seconds_batch": 27.221027806052007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4413812160491943, "target_z": 0.6093749487605703, "ce": 6.2767791748046875, "genuineness": 4.193545341491699, "identity": 0.4711324870586395, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7e6df2cafbac04cbf4e3627e5bdfa3b7f179d61889a2ab14d2b2e4ec91e484fa.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "7e7810bc82ccf36b90a9e3c45bba19685f880be9f61002328f36721579f6d162", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "58562268e6f57f17907016068461abcd52cfbe200396ff071aab7d35b322350c", "generation_seconds_batch": 22.983800269081257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0984532833099365, "target_z": 0.33984378989355596, "ce": 6.268836498260498, "genuineness": 4.111956596374512, "identity": 0.28683149814605713, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7e7810bc82ccf36b90a9e3c45bba19685f880be9f61002328f36721579f6d162.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "7e79bc6963a49c07541ca8fd50a1c2820cd1372a799472bad605121c69d0d7a9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "53cca86aa5e5e6ed7ef9cd303281348c2df224a37a3b2f086707bf9586a1ba51", "generation_seconds_batch": 17.333970076055266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.832973480224609, "target_z": 1.7031250922309735, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.8289523720741272, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7e79bc6963a49c07541ca8fd50a1c2820cd1372a799472bad605121c69d0d7a9.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "7e9959a160e7b19b26fb7c81f505d289a17fad3be56adea21eaa60cbbb79088d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ecad237de7b172beedeadfd53a114a42f444beeb252e4fe371ea971552080080", "generation_seconds_batch": 20.20499425707385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 5.998785495758057, "genuineness": 4.381826400756836, "identity": 0.459687203168869, "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_S_MONO_reg/7e9959a160e7b19b26fb7c81f505d289a17fad3be56adea21eaa60cbbb79088d.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "7e9ed9a99a50e1f778b8a2083e37fa48376ff6f17b1b36e2bf74a4d0cb65c4ac", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "279d0a1b2966b6cc58f7b089886f2ce9a1d5cea3033063a02d2d3069a662092a", "generation_seconds_batch": 23.50982217793353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242582321166992, "target_z": 0.45312492826479844, "ce": 6.070269584655762, "genuineness": 4.720732688903809, "identity": 0.6506749391555786, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7e9ed9a99a50e1f778b8a2083e37fa48376ff6f17b1b36e2bf74a4d0cb65c4ac.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "7edf73ae8962e1419aae13bafdc9cdde7305a5c1f1214bfaea478bdc95e3ad16", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bf05723c3e650b911c86514d086757f25648d8a2389bfc63b5b32af4b6850e64", "generation_seconds_batch": 24.067998295999132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6352100372314453, "target_z": 0.7617186391032343, "ce": 6.086154937744141, "genuineness": 4.432034492492676, "identity": 0.5526512861251831, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7edf73ae8962e1419aae13bafdc9cdde7305a5c1f1214bfaea478bdc95e3ad16.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "7f0fe43658c0c08fa33ac8d0fe72d679066609aa01b798989e96a69ba06baa9b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "72e0e8d85ea7ff148574b8f1fde2659a7f63761eaf9af7ef6f8ea080f34d04bc", "generation_seconds_batch": 27.51203015097417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8389790058135986, "target_z": 0.921874989752114, "ce": 6.3641486167907715, "genuineness": 4.243753433227539, "identity": 0.5641352534294128, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7f0fe43658c0c08fa33ac8d0fe72d679066609aa01b798989e96a69ba06baa9b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "7f25542896c95a1aff64e9685fa67ab67677adfed03ba3d0dc0b7ecbef4cf62a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26d8bd594985c0fe0dad1131c3a71dea52e10635e81a95831777031dc825517d", "generation_seconds_batch": 21.34215405106079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 5.704906940460205, "genuineness": 4.444586753845215, "identity": 0.7520418763160706, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/7f25542896c95a1aff64e9685fa67ab67677adfed03ba3d0dc0b7ecbef4cf62a.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "7f650c1c8722ed2c3bd554694a8abbbbe2e5bfb00ac7a4fe8d74670731059844", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aba725755cfd21f957b45c9de03cb1423d4b65a7a958b9fdc40c57e0ac7a719d", "generation_seconds_batch": 27.811964399996214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8290390968322754, "target_z": 0.9140625168358126, "ce": 6.24500846862793, "genuineness": 4.758388519287109, "identity": 0.7602511644363403, "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_S_MONO_reg/7f650c1c8722ed2c3bd554694a8abbbbe2e5bfb00ac7a4fe8d74670731059844.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "26fee7b722af46a833fdb04eabc93668d8193f3f5dd90a7cc6f59ea3982c01e5", "generation_seconds_batch": 17.752049432019703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8835020065307617, "target_z": 0.17089847748505496, "ce": 6.395919322967529, "genuineness": 4.425758361816406, "identity": 0.219915509223938, "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_S_MONO_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "8061994221445bc143772c41f0a7e116ca78c1fdd1e46fe361fb929440aade17", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f712e4fba18f3ac48d1cf6a1e41ebd90cd1e89025d1d2db8b4e787329b00270b", "generation_seconds_batch": 27.399289103923365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.127237319946289, "target_z": 1.1484374538845132, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.7212530374526978, "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_S_MONO_reg/8061994221445bc143772c41f0a7e116ca78c1fdd1e46fe361fb929440aade17.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "80abf171a5b7cde5623a387880ada5e6ca8211fbb9e9ae72a5d4a8259defa508", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e1bf5da0107d968647d1f89ee765ec842bba48b8bc58609372aa295b8f4f471f", "generation_seconds_batch": 22.07985139603261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.205295085906982, "genuineness": 3.8922953605651855, "identity": 0.42902320623397827, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/80abf171a5b7cde5623a387880ada5e6ca8211fbb9e9ae72a5d4a8259defa508.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "13b36d1d2036e337db27759376c03171732d8dbabee2d132ddcd47d375e03ac4", "generation_seconds_batch": 16.65727922297083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9582583904266357, "target_z": 1.0156250395275601, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.48792654275894165, "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_S_MONO_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "8132baab5807376bf43b8494305268922f893cb820dc9424cf901dd266742ee0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "43fe01c0a9069d921858441698a213ccde4eb971e317732eb11bf43ec52a2a21", "generation_seconds_batch": 17.89656670903787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.268836498260498, "genuineness": 4.319066047668457, "identity": 0.7122415900230408, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8132baab5807376bf43b8494305268922f893cb820dc9424cf901dd266742ee0.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "81ddd5a40edf7c00f0018ef9331b8627d1b0de46831e89f80a3f176789bc768a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "545e683b5335ce817e7ba87347cc249198b0bbbf3b9a8a093a3237d2c688d0f8", "generation_seconds_batch": 22.62317726504989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5556907653808594, "target_z": 0.6992188557728227, "ce": 6.038498878479004, "genuineness": 4.406930446624756, "identity": 0.5301207304000854, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/81ddd5a40edf7c00f0018ef9331b8627d1b0de46831e89f80a3f176789bc768a.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "8202997accf489c6ebdb6930e902a8be297cc9edba3029fbef2636f24fa0da75", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d7717eef855dbe72719cbb33adee25a0db3f1f6a9da7312204a80dd2684f1bbf", "generation_seconds_batch": 16.82530953292735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.340320587158203, "genuineness": 4.20609712600708, "identity": 0.49878811836242676, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8202997accf489c6ebdb6930e902a8be297cc9edba3029fbef2636f24fa0da75.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "8212f880c1cb230ffb720c9177e9685eaa9c1f9acc4215849c0e86738fd6aca6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3ca0827254a156fc4de33fa38ed5e168a0a30dea2f8a64e0b1e5de68013d0566", "generation_seconds_batch": 23.183315658941865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1332430839538574, "target_z": 0.3671875387955682, "ce": 6.205295085906982, "genuineness": 4.256305694580078, "identity": 0.7538703680038452, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8212f880c1cb230ffb720c9177e9685eaa9c1f9acc4215849c0e86738fd6aca6.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "821eeaac93024ddd46911cedc561d0dcd52746263c25cc66023a1427b62f4f1d", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91b5c90b272048e5d370ccd479124236dcf7d9a86513e486110caa514ab9491e", "generation_seconds_batch": 19.921126255998388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.107357501983643, "target_z": 1.1328125080519105, "ce": 6.165581703186035, "genuineness": 4.149612903594971, "identity": 0.5261868238449097, "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_S_MONO_reg/821eeaac93024ddd46911cedc561d0dcd52746263c25cc66023a1427b62f4f1d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "82873f4e885383208af000b5b039c09aa9656b0d7c3f28c70b66ad441dcb6c81", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0d462d4f2ca9cfc1ede5ab8443c8979f19103762bc37cd6a35fdb95669a064e4", "generation_seconds_batch": 22.65118133602664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.505990982055664, "target_z": 0.6601563038014012, "ce": 6.292664527893066, "genuineness": 4.093128204345703, "identity": 0.4779796302318573, "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_S_MONO_reg/82873f4e885383208af000b5b039c09aa9656b0d7c3f28c70b66ad441dcb6c81.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "82f398bd15547ff733c03b7867084df1ec9a3c48b1e0b32ec0ec4d1f62df76c1", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d4def692a092b68b8ea9fc080d271f9aabc6f5b3f2fedacd2c88e59a5cf2056a", "generation_seconds_batch": 17.21346308698412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 6.125868320465088, "genuineness": 4.406930446624756, "identity": 0.7469727993011475, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/82f398bd15547ff733c03b7867084df1ec9a3c48b1e0b32ec0ec4d1f62df76c1.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "82fc6ba72cbc2d368cbb335dfbaa980b6897d7d9423b958336fac991e96011da", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "924605164eabf395c0fae45d75471bd6380eb90d822c83503d081d6d0d9c6493", "generation_seconds_batch": 18.435334975016303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0760884284973145, "target_z": 0.3222656789843991, "ce": 6.173524379730225, "genuineness": 4.243753433227539, "identity": 0.6265645623207092, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/82fc6ba72cbc2d368cbb335dfbaa980b6897d7d9423b958336fac991e96011da.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "8332fd0b4363040259ac411f3a5f559ca97c9914739f9f0bdfc80a2f5e960f63", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a023d5c3c0665b2d974381ccc8d1bc4e3b937b87e0b7579519ca919e8813daa9", "generation_seconds_batch": 23.765001379069872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.959294080734253, "target_z": 0.2304687942855071, "ce": 6.24500846862793, "genuineness": 4.231201171875, "identity": 0.35475096106529236, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8332fd0b4363040259ac411f3a5f559ca97c9914739f9f0bdfc80a2f5e960f63.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "837f5b7988fe526d4f5f6b234b378b39cb0975af7e82b30cb9446b0cc2a840eb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7327b1750ce0aae91c45d78d0c683040ec57f234a2798871a634fd550788596b", "generation_seconds_batch": 18.14616575895343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5954504013061523, "target_z": 0.7304687474380285, "ce": 5.736677646636963, "genuineness": 3.9299514293670654, "identity": 0.6611046195030212, "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_S_MONO_reg/837f5b7988fe526d4f5f6b234b378b39cb0975af7e82b30cb9446b0cc2a840eb.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "8395585ddf68f635d9d8e136922622c260070ac3f68313870e741d14e00af5c8", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "425d0e63fd27b693e7c2004c45b26946f0d824c8f3be9fca6060ada990c0ea96", "generation_seconds_batch": 20.497960462933406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 6.308549880981445, "genuineness": 4.557555198669434, "identity": 0.7701472043991089, "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_S_MONO_reg/8395585ddf68f635d9d8e136922622c260070ac3f68313870e741d14e00af5c8.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "8520b23482bff62396a2f3dab9874c3a92a05a96d0f71531324e32bd32ddbc93", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c4b25441fe84dab4dbcc2fe4c7e49e11870df2b08ad3870f244d90c0f3c1a2aa", "generation_seconds_batch": 16.952595390030183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3718016147613525, "target_z": 0.5546874509565459, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.7158126831054688, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8520b23482bff62396a2f3dab9874c3a92a05a96d0f71531324e32bd32ddbc93.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "852f014f72d2dabbb997677e7d0894120b5885d6cfb96323d77f765038a69d17", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "868e2ab370c24040a0af7ac9d549d882a85fe184b9af648ce542ad517510bd22", "generation_seconds_batch": 16.64903970900923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824069023132324, "target_z": 0.9101561866827047, "ce": 6.086154937744141, "genuineness": 4.15588903427124, "identity": 0.7072421908378601, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/852f014f72d2dabbb997677e7d0894120b5885d6cfb96323d77f765038a69d17.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "8534a175a636067f4f6194a95477005aabfbc874b0a1c731552b66e7059587ff", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a18f66b4bb062b5378f6558685d5511e657867a98fd16c0fe5479ad16dcbd290", "generation_seconds_batch": 16.72134528902825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.467403411865234, "genuineness": 4.344170093536377, "identity": 0.45105165243148804, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8534a175a636067f4f6194a95477005aabfbc874b0a1c731552b66e7059587ff.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "85b7c162f9d474f59be645c86bceed6762e9ebcc6f5eec037f1835eddab457e1", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f3c1fb10bef659bdccc1de9409e1a32808db210b06fa69211851ff3ae6542216", "generation_seconds_batch": 17.61940596299246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.505990982055664, "target_z": 0.6601563038014012, "ce": 6.38003396987915, "genuineness": 4.231201171875, "identity": 0.5238460898399353, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/85b7c162f9d474f59be645c86bceed6762e9ebcc6f5eec037f1835eddab457e1.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "85e0b4c645378a4e1c459d13f7c33ffcc5bd2e568609a9d1a4c016e5cb3227df", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9d0ba38526a735b93b89d598dd7314415beec566dbde87bb423580f6d2629fbd", "generation_seconds_batch": 16.342151918099262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.774162530899048, "target_z": 0.08496090062591043, "ce": 6.2767791748046875, "genuineness": 4.482243061065674, "identity": 0.5592296123504639, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/85e0b4c645378a4e1c459d13f7c33ffcc5bd2e568609a9d1a4c016e5cb3227df.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "85e3bf20c4bb4b14fb2e53984b9461196d99adbbd8a3e00aa34641a00df90bf6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5f67f722db57324a91de7486331fe6b55e3ef757d276c5da5e1e24a9e739f9ec", "generation_seconds_batch": 25.644426582963206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.031772136688232, "target_z": 1.859374925336831, "ce": 5.982900142669678, "genuineness": 4.243753433227539, "identity": 0.6524355411529541, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/85e3bf20c4bb4b14fb2e53984b9461196d99adbbd8a3e00aa34641a00df90bf6.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "85eb5a90dd0c050579516c3159c85dec96dd89ee9d17359096514cd2359ec91a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ed096847fc947ec53ab59fd951d5eccade8344afa1a016f8ff2b46b6154f0292", "generation_seconds_batch": 24.745266709011048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2525224685668945, "target_z": 0.4609375885710142, "ce": 6.070269584655762, "genuineness": 4.394378662109375, "identity": 0.6543452739715576, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/85eb5a90dd0c050579516c3159c85dec96dd89ee9d17359096514cd2359ec91a.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "85fd4a6462973ffada4b68603d5985daa769d7344fbd06a0f2fd13105795667f", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ab75000d6820811357ab5aa4b0ad2f207537e66cc6c5b0eec437dc4aec5fa25", "generation_seconds_batch": 27.677831275970675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.654054641723633, "target_z": 1.5625002049577188, "ce": 6.062326908111572, "genuineness": 4.3881025314331055, "identity": 0.8718575239181519, "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_S_MONO_reg/85fd4a6462973ffada4b68603d5985daa769d7344fbd06a0f2fd13105795667f.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "8627d1829ce23bb846f0656cda94cbdb1562ef309ca39e600c1b84e2086f27b9", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6914493778ac251a77f8da3e75d8fbb94f9d36aefd6642ff9404bd478b4353f2", "generation_seconds_batch": 17.814961828989908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 5.871703147888184, "genuineness": 4.482243061065674, "identity": 0.7108570337295532, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8627d1829ce23bb846f0656cda94cbdb1562ef309ca39e600c1b84e2086f27b9.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "868004e88d3fa48ea8ce17b43c6675381708f8252fd584aa4a20debd499f9014", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0580c5ea991bb4ac4fb1a08a4e9c52a803ce830a080bf2f4740dbd81d70cecf1", "generation_seconds_batch": 22.783045891905203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 5.736677646636963, "genuineness": 4.055472373962402, "identity": 0.5809075832366943, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/868004e88d3fa48ea8ce17b43c6675381708f8252fd584aa4a20debd499f9014.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "86aa543f88a18f23868aa522985acb71421545b2f0e63f82d0334e1af2e5e444", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c7e9648278f2a530fc32d143b3d249fcbda01fd5ff8fe67d78b9d26d52230eb4", "generation_seconds_batch": 26.512967976974323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 5.974957466125488, "genuineness": 4.494794845581055, "identity": 0.8062597513198853, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/86aa543f88a18f23868aa522985acb71421545b2f0e63f82d0334e1af2e5e444.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "86eef4804c8d6fe97d1dbb9d60f87119b00cefac8cb7912cc0c61df2174fbdb5", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "160080c3b426d44e6f9a6e4961638abbc8b55817a10c43407a0e553120592c7e", "generation_seconds_batch": 20.55550511495676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 6.316492557525635, "genuineness": 4.7081804275512695, "identity": 0.75971519947052, "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_S_MONO_reg/86eef4804c8d6fe97d1dbb9d60f87119b00cefac8cb7912cc0c61df2174fbdb5.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "87542a3ea3da0ca13635245adab73e8e07558b99d0bbb5e23052279baebd805e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ccbbaf07dacdc17e498ec674e17aa868b5aed7fd996723866b61540739973344", "generation_seconds_batch": 17.674420837080106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6352100372314453, "target_z": 0.7617186391032343, "ce": 6.459460258483887, "genuineness": 4.557555198669434, "identity": 0.2501266896724701, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/87542a3ea3da0ca13635245adab73e8e07558b99d0bbb5e23052279baebd805e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "87a53d6dd6d5200af47ddc1e316b5f4bc6ce672578c9a749c3dd5c2a0316226a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b09a3b19e8c1bc9ca9048065a25d0e94a0ad7445ca6a4717ec2d302a11191cd6", "generation_seconds_batch": 18.03129886800889, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7942492961883545, "target_z": 0.8867187679338004, "ce": 5.7406487464904785, "genuineness": 4.030367851257324, "identity": 0.5664935111999512, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/87a53d6dd6d5200af47ddc1e316b5f4bc6ce672578c9a749c3dd5c2a0316226a.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "88612399e3a14d4bc867557fd0ee38739743e5da0c26c39cd10dccd4d61e640e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8976a7de17dbe3dd84a5334be91d965a3d7414d3cfb5652b7d51ba8697cda29c", "generation_seconds_batch": 27.45250523800496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.673933982849121, "target_z": 1.578124776010493, "ce": 6.157639026641846, "genuineness": 4.519899368286133, "identity": 0.5056788325309753, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam neem ze mee als je wilt.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/88612399e3a14d4bc867557fd0ee38739743e5da0c26c39cd10dccd4d61e640e.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "88a2d3c925558ce65aadf746fea99a4d79d853bd7c2d65e284c0abd2b70ac640", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5ed0492664d0f1d63b7856bb36b35cd21a61c28dd20ad0bc603c22e779ab30b0", "generation_seconds_batch": 22.943207517033443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.704789638519287, "target_z": 0.8164061369072587, "ce": 6.038498878479004, "genuineness": 4.494794845581055, "identity": 0.6759219169616699, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/88a2d3c925558ce65aadf746fea99a4d79d853bd7c2d65e284c0abd2b70ac640.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "88b525c934b309aa85071b799fe1590dac092354f0a00903bd2c780080352303", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d5fac61810d60ab3eb1b3842a8c692ab59975134c8531cc03d5eb2a1376d0eac", "generation_seconds_batch": 28.152758267940953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.6838222742080688, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/88b525c934b309aa85071b799fe1590dac092354f0a00903bd2c780080352303.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "88cc03db65543ed163e9a79181951967511b08208cbde23c08af54b5c430540c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7e9be11d4721cdd982f4d7b6582cd2be7c7a412709fec3482c94900e52eaa6a9", "generation_seconds_batch": 21.29275748692453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.268836498260498, "genuineness": 4.162164688110352, "identity": 0.24486826360225677, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/88cc03db65543ed163e9a79181951967511b08208cbde23c08af54b5c430540c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "8904a7a1a3b1ea09f2b6ea5273a162b42f2665cf4f492db779638fb780acedda", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a1872238da65ecdc575badb283f444978edc5f20d3c22673e11a0df1d609b220", "generation_seconds_batch": 21.34215405106079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.575570583343506, "target_z": 0.7148438016054256, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.7144787311553955, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8904a7a1a3b1ea09f2b6ea5273a162b42f2665cf4f492db779638fb780acedda.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "8943c2bccb01248ce39691e079da54dac6b57fab0a983aad2c08f794a7d1ef36", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d83b65012a9c2bd80aac43a129fe0cee1bd994fd070ee7ce90c758c61a0f4b37", "generation_seconds_batch": 16.88917155202944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0760884284973145, "target_z": 0.3222656789843991, "ce": 6.1020402908325195, "genuineness": 4.331617832183838, "identity": 0.7976211309432983, "hypothesis": "De deur 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_S_MONO_reg/8943c2bccb01248ce39691e079da54dac6b57fab0a983aad2c08f794a7d1ef36.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "894c8333f0ad41aa37cd182db9cc27772c5ce746036af4e209af1f2a64532c72", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "80e994888cfbb620b5b1fcadfb7a1cfb3239b1413456b3258bfa13201fef1ae2", "generation_seconds_batch": 23.962809504009783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.535810708999634, "target_z": 0.6835937225503055, "ce": 6.181467056274414, "genuineness": 4.65797233581543, "identity": 0.675559401512146, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/894c8333f0ad41aa37cd182db9cc27772c5ce746036af4e209af1f2a64532c72.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "895cc62fa756bdef43f63f563afaff47487d121c0bd4a8c656ba8d740ffef3fe", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "28487ad9064706a1e1f72d0c8a5afecb1cee3a3fd3793e6d08561cfa13b973ed", "generation_seconds_batch": 23.362556381034665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.086154937744141, "genuineness": 4.582659721374512, "identity": 0.6539943218231201, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/895cc62fa756bdef43f63f563afaff47487d121c0bd4a8c656ba8d740ffef3fe.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "8972b8f98c3d1ba5a3a83734346c3a41aa2951dca1593729922643220ff5aceb", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "171aa3b04ec436de5533484fdcd884762fc30548dcd0a1ae61bf1990a0457ccc", "generation_seconds_batch": 20.547841383027844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.998018264770508, "target_z": 1.0468751185826801, "ce": 6.24500846862793, "genuineness": 4.369274139404297, "identity": 0.6967124938964844, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8972b8f98c3d1ba5a3a83734346c3a41aa2951dca1593729922643220ff5aceb.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "89b53723b2f0411ad00e924f3a8e2e9ee6ac175153362a1e158281e00ad0dc7c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "892e4f71239b9b8c9f36594297b30356dd45bd666f8456022c05df890c1fa364", "generation_seconds_batch": 16.868734539020807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.41180419921875, "genuineness": 4.419482707977295, "identity": 0.30617737770080566, "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_S_MONO_reg/89b53723b2f0411ad00e924f3a8e2e9ee6ac175153362a1e158281e00ad0dc7c.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "89c19c1870d9a696130e97580de8e1e7fd37583c79902f7066e40e2572f59985", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "373b5eb10e418049414041d1e317a69f57074a2f4741737ede045e4feb4dc58d", "generation_seconds_batch": 21.893815565039404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.173524379730225, "genuineness": 4.375550270080566, "identity": 0.43705081939697266, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/89c19c1870d9a696130e97580de8e1e7fd37583c79902f7066e40e2572f59985.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "89c5bb07a9f6413bf41ee41f94829934eb4bf52731271953e68e1e126799d04f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "38e228ae0715673875da8f1ba67ce5c6ad92169fc693631b23b847cc725560a7", "generation_seconds_batch": 24.418124006013386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.125868320465088, "genuineness": 4.375550270080566, "identity": 0.4843556582927704, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "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_S_MONO_reg/89c5bb07a9f6413bf41ee41f94829934eb4bf52731271953e68e1e126799d04f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "89d7a930b642df1f6d8a0c974a8a0364bd325f02c5671470b6f2df4e09a42dd6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0aa637a04c1334eef867c1895df120da520dda79401fdb90655775eea049bf68", "generation_seconds_batch": 16.18322566500865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.5750771760940552, "hypothesis": "De deur 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_S_MONO_reg/89d7a930b642df1f6d8a0c974a8a0364bd325f02c5671470b6f2df4e09a42dd6.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "89d9dc555d598f2ed2025512249aa10984ddb7a8f968b4990fb49ba006b03f9a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5047ce6c82a1c11670e242ca18075dd287701d74b7e1d3aba4102db74835b4a6", "generation_seconds_batch": 16.68167663703207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.416531562805176, "target_z": 0.5898438601647739, "ce": 6.260893821716309, "genuineness": 4.3127899169921875, "identity": 0.7688920497894287, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/89d9dc555d598f2ed2025512249aa10984ddb7a8f968b4990fb49ba006b03f9a.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "89daf8452cbc4ef5a9f823135746eb8db0c2e53604f2da9556d603763fffce68", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8a299d1985a19c04ec4b5d9755e5c506894dfa427fa91c9f0e1fc9379f55512a", "generation_seconds_batch": 23.613810019916855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1471171379089355, "target_z": 1.1640623997171162, "ce": 6.109982967376709, "genuineness": 4.319066047668457, "identity": 0.4200938940048218, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/89daf8452cbc4ef5a9f823135746eb8db0c2e53604f2da9556d603763fffce68.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "89ea670c4fb695db6e783704d1fa732f8c14eb304d2451ac3550a2baaf8d723b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7f6c520dc81c879970b56d6a47db9afe9698f43664bf6f7efea675004a8e8cf2", "generation_seconds_batch": 21.108577296021394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 5.974957466125488, "genuineness": 4.65797233581543, "identity": 0.6210207939147949, "hypothesis": "La porte 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_S_MONO_reg/89ea670c4fb695db6e783704d1fa732f8c14eb304d2451ac3550a2baaf8d723b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "8a15211e42ba301a4f1b7b35f76ce22655244d9a034732b7af702ddff9fe8f2e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7f63caa40c99a9a074234571aee0d962f0a3119742e2f364374cb218322bf050", "generation_seconds_batch": 27.298380355001427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.1020402908325195, "genuineness": 4.519899368286133, "identity": 0.5725209712982178, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8a15211e42ba301a4f1b7b35f76ce22655244d9a034732b7af702ddff9fe8f2e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8a2791f3f8d27860ca17b3c85d8dbcc1755ec0fef9fad94cee7a6c9d91131022", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4235abda04c31598d53f0fceffd21bef02b6e65c4d5d8ebdd23d11a4b6875d44", "generation_seconds_batch": 22.673166669905186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8290390968322754, "target_z": 0.9140625168358126, "ce": 5.8875885009765625, "genuineness": 4.231201171875, "identity": 0.6018628478050232, "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_S_MONO_reg/8a2791f3f8d27860ca17b3c85d8dbcc1755ec0fef9fad94cee7a6c9d91131022.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "8a29aded40338d342ee7d7adfada3d46efe0637048b01e02de34b67e3738f72c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0aea713a85cbcaaf547fcafe1d4086b8e340806d18057d8fcd5286e23de2f845", "generation_seconds_batch": 23.253210914088413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.356891632080078, "target_z": 0.5429686478871365, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.7799185514450073, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8a29aded40338d342ee7d7adfada3d46efe0637048b01e02de34b67e3738f72c.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "8a46a122bd4a91ecc4d272cf011e182a93fcff7689c15d5efd98a77dbeec05c5", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "284f6bcb0d9f31cd9a001806836482ad08da1bfead86098e3060a45fe1582cc2", "generation_seconds_batch": 16.987685942905955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2748870849609375, "target_z": 0.4785155120902567, "ce": 6.419747352600098, "genuineness": 4.6203155517578125, "identity": 0.44568800926208496, "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_S_MONO_reg/8a46a122bd4a91ecc4d272cf011e182a93fcff7689c15d5efd98a77dbeec05c5.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "8a5990d4f91094468f03a81fa5464ce86c15f44675eb9a903b804caaaf7d54cc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fa9a0cb861e2757064bf273fd474f2c63135e665bc269c160251a06d8fbabd96", "generation_seconds_batch": 18.389203201048076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 6.24500846862793, "genuineness": 4.193545341491699, "identity": 0.3308500349521637, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8a5990d4f91094468f03a81fa5464ce86c15f44675eb9a903b804caaaf7d54cc.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "8a6024581c32324e5a29ee334d99681eb9f8d4e50a0192f61882255a3e5cc20f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0590b5e2e075725cb9316682d33f6044d4118374876ee1f104de124b7f697499", "generation_seconds_batch": 23.811156190000474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246516704559326, "target_z": 1.2421875036599592, "ce": 5.77639102935791, "genuineness": 4.532451152801514, "identity": 0.577138364315033, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8a6024581c32324e5a29ee334d99681eb9f8d4e50a0192f61882255a3e5cc20f.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "8af2ee91437f01e19eb6d60bf8ac01447e6f61e9a010153a285e960d3f58002d", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09514f31a8d078d1dae6cab0267a21f0d72e690167924ff4c981ae464f0b3317", "generation_seconds_batch": 27.56565720506478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.594414710998535, "target_z": 1.5156249926800816, "ce": 6.268836498260498, "genuineness": 4.118232727050781, "identity": 0.4765094518661499, "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_S_MONO_reg/8af2ee91437f01e19eb6d60bf8ac01447e6f61e9a010153a285e960d3f58002d.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "8b6af8cd53d6a95ebeb9bbc5c727c9d3b7972a9369d5d0d2635f143b90ce4757", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "038cca076d99448d156e3a1390ec4f9301616f1f6f665bda121550eadb39f1e0", "generation_seconds_batch": 22.64624836598523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8439488410949707, "target_z": 0.9257811325153077, "ce": 5.959072113037109, "genuineness": 4.331617832183838, "identity": 0.6220506429672241, "hypothesis": "De deur 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_S_MONO_reg/8b6af8cd53d6a95ebeb9bbc5c727c9d3b7972a9369d5d0d2635f143b90ce4757.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "8ba159e7fed5c721646a936aed05213f779f9b8ded5a63f6260ed72e0abbb6fb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bce3a0fb48303be1f2b8e19ba341405c70574a5f739e69af449b7bd879a6b231", "generation_seconds_batch": 23.613810019916855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.213237762451172, "genuineness": 4.3881025314331055, "identity": 0.6965949535369873, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8ba159e7fed5c721646a936aed05213f779f9b8ded5a63f6260ed72e0abbb6fb.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8bb1dfd62a736eece139cf4a8b3569839faf4f3d1940bc6a08ae443c716ff093", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "43b9e48c7ba0b41ff487e8510f79518c928601719664899649f80450bd709d72", "generation_seconds_batch": 24.81252883700654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 5.77639102935791, "genuineness": 4.419482707977295, "identity": 0.762803316116333, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8bb1dfd62a736eece139cf4a8b3569839faf4f3d1940bc6a08ae443c716ff093.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "8bbe5c05c016dac5da492f15c05b162384bfec6d0b7d512838844f6d450f26ee", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "182412a27c5ffcc5fa51f55d652765d3da52dcd953923744d184f9438f32da0c", "generation_seconds_batch": 16.96470561006572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.1894097328186035, "genuineness": 4.595211505889893, "identity": 0.6187001466751099, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8bbe5c05c016dac5da492f15c05b162384bfec6d0b7d512838844f6d450f26ee.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "8c1641bcc3f8950fd7e2cb09e0c78d71a6a7a3423c3fb8cb90274675886e80c2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "486c07e27cda772003db23ab4dca0adbba49597cd54c512a6092611ba8b2d588", "generation_seconds_batch": 26.411029117065482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.054384231567383, "genuineness": 4.344170093536377, "identity": 0.8189988732337952, "hypothesis": "La porte 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_S_MONO_reg/8c1641bcc3f8950fd7e2cb09e0c78d71a6a7a3423c3fb8cb90274675886e80c2.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "8c1f4a74eb47e5d9f7e1a30e23ea457c36c43bec1fb1bdadd20e421614868909", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d65d719d59f4bf2936fa62f76191d16c5f9db7fa4d16f24946ff2426b42c86f1", "generation_seconds_batch": 16.938079711981118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.216696739196777, "target_z": 1.2187498975211406, "ce": 6.197352409362793, "genuineness": 4.507347106933594, "identity": 0.6533318758010864, "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_S_MONO_reg/8c1f4a74eb47e5d9f7e1a30e23ea457c36c43bec1fb1bdadd20e421614868909.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "8c23bd3124422e04a53106f2b1639a1a26daba3f56fcf26ad6dac735cffa8751", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d6c9aabbf39bc8c59716872efb653e0586833e5d4282bcf752a0ae688a424201", "generation_seconds_batch": 22.669119808939286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.070269584655762, "genuineness": 4.1998209953308105, "identity": 0.5714430212974548, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8c23bd3124422e04a53106f2b1639a1a26daba3f56fcf26ad6dac735cffa8751.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "8c3a8b25f51b3f0df772d65d0c29e41d68a5c9de2204c62b3ce61350f82618f2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "33c47f1a40db8a7ca17babead1255a284258d3ed3b59e737b0e53037decac537", "generation_seconds_batch": 17.596382108982652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.633760929107666, "target_z": -0.025390583825458286, "ce": 6.300607204437256, "genuineness": 4.168440818786621, "identity": 0.23367957770824432, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8c3a8b25f51b3f0df772d65d0c29e41d68a5c9de2204c62b3ce61350f82618f2.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "8cfb17e7f15ecdfb1a89bc532d490d4c5654f4b1edc8dde37800ed040ee3ed19", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "628fa02f4b93ade5320cd9fb2708a70525a4cb4e877967ecc35362b4467f517d", "generation_seconds_batch": 17.594369477941655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.814129114151001, "target_z": 0.9023437137664033, "ce": 6.173524379730225, "genuineness": 4.758388519287109, "identity": 0.6882333755493164, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8cfb17e7f15ecdfb1a89bc532d490d4c5654f4b1edc8dde37800ed040ee3ed19.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "8cfe635cd7fa7c41ae4f1c882222481b457ee078fbb3b9683ed5aecabd893ad4", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bcd973002d90d66e6b518f31dee8348fc43f392f9a3115faffbe81a7f128cec", "generation_seconds_batch": 16.903253169031814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.141753673553467, "genuineness": 4.758388519287109, "identity": 0.8183892965316772, "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_S_MONO_reg/8cfe635cd7fa7c41ae4f1c882222481b457ee078fbb3b9683ed5aecabd893ad4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "8d2e9adf077a34433bcb1960f4e6700f83fca2366d455c6167ed77d4e2bc005a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c4339bf24066ae53dd6db1dc8c4fc4fdc6d7e6b1f9c7d3bd1d2ad3273b3dfa0f", "generation_seconds_batch": 23.794424440013245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.086154937744141, "genuineness": 4.413206577301025, "identity": 0.5551365613937378, "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_S_MONO_reg/8d2e9adf077a34433bcb1960f4e6700f83fca2366d455c6167ed77d4e2bc005a.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8d7ed383c10b1724f440323c51b8be829e1d6be1438db146ebc257f106c0adc4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54db1f60161aaee45557cacbf177e575500f08dbd7f07c3244ff98e40f62e315", "generation_seconds_batch": 17.11183617694769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9880781173706055, "target_z": 1.0390624582764645, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.5761264562606812, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8d7ed383c10b1724f440323c51b8be829e1d6be1438db146ebc257f106c0adc4.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8dd87c525170b8a6b27f655099ef7c15a701e3e162cad82f1cd551a36c5c940b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "469393f01f704dcbba614b4fd1ef2ddd55487739dcd268c460ebb3da35e7aab8", "generation_seconds_batch": 21.70670681493357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6103603839874268, "target_z": 0.7421875505074378, "ce": 5.895530700683594, "genuineness": 4.124508857727051, "identity": 0.38978955149650574, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8dd87c525170b8a6b27f655099ef7c15a701e3e162cad82f1cd551a36c5c940b.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "8e96f4a41d45be142bc6f61b846aae4cc0d2e6d8735d9e2bbaf50fed9628e3c0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fdba404c529f5073aafc1b0acf64db470ac81bc1e4597ee3991b203c6ef1f07c", "generation_seconds_batch": 24.572302518063225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7644295692443848, "target_z": 0.8632813491848961, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.842766284942627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8e96f4a41d45be142bc6f61b846aae4cc0d2e6d8735d9e2bbaf50fed9628e3c0.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8ea8e12f3887854be684d1f4cd07459c22397c977fc681ae1fd799659b87a6c4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de874d57fef5d74b83eeddb187e3645f38b72f33870d28610481286f32609efa", "generation_seconds_batch": 24.163314861012623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.753453731536865, "target_z": 1.6406249341207333, "ce": 5.202532768249512, "genuineness": 4.105680465698242, "identity": 0.46828633546829224, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo está.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8ea8e12f3887854be684d1f4cd07459c22397c977fc681ae1fd799659b87a6c4.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "8eb156647e47f3f1cd21a08cb1ec39923b77d3749f9218c679e37ea20cc2bb90", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b8d4688f12d7444551210bbd8afb646abdeecdb00c49d76df1532a84103f9d2a", "generation_seconds_batch": 17.924006558954716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 6.109982967376709, "genuineness": 4.645420074462891, "identity": 0.7485696077346802, "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_S_MONO_reg/8eb156647e47f3f1cd21a08cb1ec39923b77d3749f9218c679e37ea20cc2bb90.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "8f243870d2d4fbf7a80b870cf3cdf6ca1e8fd4b4fbeddcbaa934fde3214cac4e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "691a05497dd53418bacf1fb174377494ef7d7d7b0dc38ac32a3dfd3192a7840d", "generation_seconds_batch": 22.49290754098911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 5.744620323181152, "genuineness": 4.168440818786621, "identity": 0.8329814672470093, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8f243870d2d4fbf7a80b870cf3cdf6ca1e8fd4b4fbeddcbaa934fde3214cac4e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "8f3e0201a68ab0e979f7c1e450348dbc2439aa5e0f643df2bf98fec790f7e785", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5790b91c11ff9c475c9c9646ae2bf90cbe3ce7721fdf9b3a94e9ad82db2c6047", "generation_seconds_batch": 17.660411798977293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8041892051696777, "target_z": 0.8945312408501018, "ce": 6.157639026641846, "genuineness": 4.174716949462891, "identity": 0.5655027627944946, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8f3e0201a68ab0e979f7c1e450348dbc2439aa5e0f643df2bf98fec790f7e785.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "8f8a5503842ef44bc7b3c885ee0edb2aa4366a929b9d7b3a81ff070218a19210", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5b785d7bdc0f2059773d04e8f95e26c3a2efe07bfd60f5c7470c6fd539789f51", "generation_seconds_batch": 22.43877272796817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9580514430999756, "target_z": 0.229492118052273, "ce": 6.062326908111572, "genuineness": 4.174716949462891, "identity": 0.5990760326385498, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/8f8a5503842ef44bc7b3c885ee0edb2aa4366a929b9d7b3a81ff070218a19210.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "9011ed8f5aad3bd30bd67fa07374e2d72336e306beb3c6f821ff2fe735010579", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "087de42ef50649737653058fccc7c2f45972d8e7f02e247b2e632d53e4506e58", "generation_seconds_batch": 23.305688432999887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 6.292664527893066, "genuineness": 4.432034492492676, "identity": 0.3874574899673462, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9011ed8f5aad3bd30bd67fa07374e2d72336e306beb3c6f821ff2fe735010579.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "906a0871b438d6a84aa5f054a2ff4840397aceac1dcf9eecead2d7410d1eac77", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5c0b7bbca84fc00307bf06b5a7f74455d2c39abe1a430772f11e2c87216e848e", "generation_seconds_batch": 21.926722283940762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7942492961883545, "target_z": 0.8867187679338004, "ce": 5.788304805755615, "genuineness": 4.356721878051758, "identity": 0.6390467286109924, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/906a0871b438d6a84aa5f054a2ff4840397aceac1dcf9eecead2d7410d1eac77.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "90bec920957c69ae424b332fb14aea70dfa2e9bdbd02113b7022dd9524379107", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "aeaa28831e74d2980b0d03b301ca11c8d21c8bf46cbdc2d15d8f00f234f92c11", "generation_seconds_batch": 22.640005494002253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.340320587158203, "genuineness": 4.356721878051758, "identity": 0.7144412994384766, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/90bec920957c69ae424b332fb14aea70dfa2e9bdbd02113b7022dd9524379107.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "919aeca4def516f5bafcbe7a2e3c90b93beb25732b7c0bd03ebd6a5d93c9d8bb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6facc217d010273d4ca8663414c0d341c32bb9fd99d6fb97cef8cee230ad3b44", "generation_seconds_batch": 23.2541243910091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9880781173706055, "target_z": 1.0390624582764645, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.5780664682388306, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/919aeca4def516f5bafcbe7a2e3c90b93beb25732b7c0bd03ebd6a5d93c9d8bb.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "91bfff09e160a0938a469f22649bd3a7b35bcff45dd421306788ff7fe743a568", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "74cb214ba398fc87258a155d56f06d331297116a83312950d4fa3ec7386fde86", "generation_seconds_batch": 28.32258645503316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7505552768707275, "target_z": 0.06640630087343376, "ce": 6.23706579208374, "genuineness": 4.758388519287109, "identity": 0.6711676716804504, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/91bfff09e160a0938a469f22649bd3a7b35bcff45dd421306788ff7fe743a568.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "937c99c89d5f97f979cae25c866595adacb1435a664ac37e19997f8787e64249", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7669751c5a22de924e789ea812ea6a37e3c060be6b18fcf7574e1bc5c9ee9089", "generation_seconds_batch": 20.514942312031053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0065088272094727, "target_z": 0.2675781811803747, "ce": 6.221180438995361, "genuineness": 4.432034492492676, "identity": 0.5567956566810608, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/937c99c89d5f97f979cae25c866595adacb1435a664ac37e19997f8787e64249.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "938b2f1fc014452a2238e2b3a848268ca07f2c3175877947e0640fa16ff7bdea", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a9a4e482b1aa30867b25b162cfd20cfee6828113e8e25bd326cfbeff5ca5d99d", "generation_seconds_batch": 23.129189759027213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 6.133810997009277, "genuineness": 4.7458367347717285, "identity": 0.5784035325050354, "hypothesis": "En que he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/938b2f1fc014452a2238e2b3a848268ca07f2c3175877947e0640fa16ff7bdea.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "93fccd40ebdd30a1d5ff7021445b572bf7f5209a708d067388804a07673c3069", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b97badd04e955fd6a1f0c65102b18dfc2c4d19aa5e9a02a7652f170bf56c7d83", "generation_seconds_batch": 16.250061376951635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7296395301818848, "target_z": 0.8359374128929695, "ce": 6.078212261199951, "genuineness": 4.281409740447998, "identity": 0.6524583101272583, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/93fccd40ebdd30a1d5ff7021445b572bf7f5209a708d067388804a07673c3069.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "940824f19cf52b5d08d1f8ff6ed569664537b203e512c78d9dd51d86a6e0a8cb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c716b7add123584aa78b07824de8973bd0509dad7ad798ccaa114059f6ff9812", "generation_seconds_batch": 24.75043791590724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.499174118041992, "genuineness": 4.632867813110352, "identity": 0.23901256918907166, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/940824f19cf52b5d08d1f8ff6ed569664537b203e512c78d9dd51d86a6e0a8cb.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "9450a4bf8199e4f32d6cfca31436eb80f051778276faa0a0083ad08860d19e97", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93c323471f4c4b277708ec5a065a588189294ad6be4f57c6a39ce16853c68604", "generation_seconds_batch": 16.958772251033224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.356205940246582, "genuineness": 4.607763767242432, "identity": 0.52653568983078, "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_S_MONO_reg/9450a4bf8199e4f32d6cfca31436eb80f051778276faa0a0083ad08860d19e97.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "94c1978ac4424bc4f01f934612c9c52d03d39b9df5cc7a0504007aa04f3b18fd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "07dee07aac9c9416b15d047f23f3af794d638a75fc8326f101494a75a748411c", "generation_seconds_batch": 27.775130321038887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.197352409362793, "genuineness": 3.9550557136535645, "identity": 0.6501106023788452, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/94c1978ac4424bc4f01f934612c9c52d03d39b9df5cc7a0504007aa04f3b18fd.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "94c8f4d520c7641ede0e559b45419ec5d5f8d1021881ee1eac314d67ff49cffb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3e593e2f2ad489476116496e6c6a46cb11d387369f65b1bd14d20c7834a1e856", "generation_seconds_batch": 24.305325884954073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.933408498764038, "target_z": 0.9960937635418493, "ce": 6.109982967376709, "genuineness": 4.2249250411987305, "identity": 0.6807758212089539, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/94c8f4d520c7641ede0e559b45419ec5d5f8d1021881ee1eac314d67ff49cffb.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "94ed5d7c7edaf2b19d92f6ccbda4d82551c7ecf6dc2036d6bf203407d5f59a89", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8df2443ea9d1cee9c9ee2399d2fc62c4bd892d5b3f41a5e5398ae8566134b801", "generation_seconds_batch": 23.069345214986242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.332377910614014, "genuineness": 4.444586753845215, "identity": 0.5305404663085938, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/94ed5d7c7edaf2b19d92f6ccbda4d82551c7ecf6dc2036d6bf203407d5f59a89.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "954f334bd1edccf62d7034aeb17227385fe3f1a40a39e383f3e348109f5dbc4e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "35b5a2ef3c1ca36f37d81f2566923d8693b377daf47c7e6898d5773755d25c7a", "generation_seconds_batch": 21.765763310017064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.173524379730225, "genuineness": 4.570107460021973, "identity": 0.5568772554397583, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/954f334bd1edccf62d7034aeb17227385fe3f1a40a39e383f3e348109f5dbc4e.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "9556d6216098cc76f668bd64c5f9a64ab8ba6d1e5b8d10674fbc6d5a1a34876e", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "efda4e0f61c2d855742818994f172efc00c3097ebf613885577e0eced33eedd3", "generation_seconds_batch": 15.991079178988002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 6.046441555023193, "genuineness": 4.118232727050781, "identity": 0.7750439643859863, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9556d6216098cc76f668bd64c5f9a64ab8ba6d1e5b8d10674fbc6d5a1a34876e.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "958579b11250925aab2fb835df94454aec0ba179918ed6f6719d8d354bcde348", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c326f2a7cd10d312f39654afd707fde70aed411045dff22b56368e8a74e3f442", "generation_seconds_batch": 24.00872556993272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3718016147613525, "target_z": 0.5546874509565459, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.2544337809085846, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/958579b11250925aab2fb835df94454aec0ba179918ed6f6719d8d354bcde348.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "95bac474f2de5c3a8ecc3a4409f099a7fbe4eb0f3c0831a354f980b0aa8a3149", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72b5265a31900c194e7e8d66aa0f916b793102216b16387d3d0079be8943bf00", "generation_seconds_batch": 17.029294381965883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824069023132324, "target_z": 0.9101561866827047, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.6164019107818604, "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_S_MONO_reg/95bac474f2de5c3a8ecc3a4409f099a7fbe4eb0f3c0831a354f980b0aa8a3149.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "95e5f32ef1a1b6371fce54e8aa68b40e219c2797b3a6b026ef1fb94071e2676e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a4f8a722490731ce573517aaae38444deb4926c2dc92b81acf688df297dbcbd", "generation_seconds_batch": 16.765714254928753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.292664527893066, "genuineness": 4.218649387359619, "identity": 0.23968231678009033, "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_S_MONO_reg/95e5f32ef1a1b6371fce54e8aa68b40e219c2797b3a6b026ef1fb94071e2676e.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "95e8b298ba2fd63d323c5722dcc8a9ae985cf36c9fc3632df99c3e1123c43482", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6643a56e47215e7a7c0c2ec1955952e393957abde6631d1e88bfd233b8badeca", "generation_seconds_batch": 29.576803146046586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9184985160827637, "target_z": 0.98437496047244, "ce": 6.173524379730225, "genuineness": 4.557555198669434, "identity": 0.751698911190033, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/95e8b298ba2fd63d323c5722dcc8a9ae985cf36c9fc3632df99c3e1123c43482.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "95fc6cd3966169d8deffd52012d3d7c70f552e9ff4e79c13ca83f0aaaf0e59d4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5ac17d461ab796d25df9c2dadc61bb49b50b1a201d87f44ed97299a53f548140", "generation_seconds_batch": 21.869271656963974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.306156158447266, "target_z": 1.289062341157768, "ce": 6.213237762451172, "genuineness": 4.18099308013916, "identity": 0.5084879994392395, "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_S_MONO_reg/95fc6cd3966169d8deffd52012d3d7c70f552e9ff4e79c13ca83f0aaaf0e59d4.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "961a3912f1412e7cc53dee5c4845427a5724a6786d96da83bda739ba6c92f5ff", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a28408089e9d835c1ab3c06dd1aa9d59a7d2fb6fea3534ce1eb58d0cef0bfb0", "generation_seconds_batch": 16.927631431026384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.882673263549805, "target_z": 1.742187644202395, "ce": 6.062326908111572, "genuineness": 4.268857479095459, "identity": 0.8264886140823364, "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_S_MONO_reg/961a3912f1412e7cc53dee5c4845427a5724a6786d96da83bda739ba6c92f5ff.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "963979ac3de7df4b8b63ae57c7877560516d27d03ed6a7e4ff75990cdf161658", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9fb9d8fbbcf7a356c380d1efadb46dbe45a41f2aef4c95f16c4eeee4f3230d6", "generation_seconds_batch": 27.68021088000387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.355855941772461, "target_z": 1.3281248931291896, "ce": 6.054384231567383, "genuineness": 4.7332844734191895, "identity": 0.7754900455474854, "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_S_MONO_reg/963979ac3de7df4b8b63ae57c7877560516d27d03ed6a7e4ff75990cdf161658.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "9640dedd24216766970e6acecc050259f75c39c4969f73b16464075c9047fbb6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7978bb9ccd437711e98043f3d44458076db673707e76d1ab3cc79f716f5fb95f", "generation_seconds_batch": 18.557513216976076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.93837833404541, "target_z": 0.9999999063050429, "ce": 6.372091293334961, "genuineness": 4.425758361816406, "identity": -0.019110247492790222, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9640dedd24216766970e6acecc050259f75c39c4969f73b16464075c9047fbb6.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "9662539d8439bb8ed20a9b56e4681955224e8d510f2b82527c8b9c080b2c3217", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e2913f8ec446f978d3f20d96b1e4ee58ebb6387d639c5969db762e36e241f0f", "generation_seconds_batch": 22.771051340037957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.814129114151001, "target_z": 0.9023437137664033, "ce": 5.95112943649292, "genuineness": 4.13078498840332, "identity": 0.5467343330383301, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9662539d8439bb8ed20a9b56e4681955224e8d510f2b82527c8b9c080b2c3217.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "9663b14c49018cf02840d6f021bbca3805c32c052b1d9b9c709a1373d9ca592d", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b362ab3e39726d41cc2aec515e4522d2c70bfcaff33755eecc27b36b2db22eb9", "generation_seconds_batch": 27.811964399996214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3359761238098145, "target_z": 1.3124999472965866, "ce": 6.23706579208374, "genuineness": 4.871357440948486, "identity": 0.7306309342384338, "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_S_MONO_reg/9663b14c49018cf02840d6f021bbca3805c32c052b1d9b9c709a1373d9ca592d.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "966a3e6a6e0447f29512cc674f58732a3592c416dcf68947f1a3862ee38d9f44", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f2ec84825b5db99243f64180f87830179b4285548fad02225d644c10b96b2c6", "generation_seconds_batch": 28.180371741997078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2962164878845215, "target_z": 1.2812500556313808, "ce": 6.292664527893066, "genuineness": 4.570107460021973, "identity": 0.5039782524108887, "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_S_MONO_reg/966a3e6a6e0447f29512cc674f58732a3592c416dcf68947f1a3862ee38d9f44.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "96aba72e816e92b54090840f1634c64f4f4a20fa0bfc851636504be2525d3fcd", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6b1d5af353492d2cf080d17b2cac544851d2869b575b7d1eae8280fcd7ea58f5", "generation_seconds_batch": 16.724924670998007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 6.41180419921875, "genuineness": 4.494794845581055, "identity": 0.2555428147315979, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/96aba72e816e92b54090840f1634c64f4f4a20fa0bfc851636504be2525d3fcd.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "96de93739b7ac206aefa380fb8755dcb6b2d1612514d30d1a3c41e6051fbf010", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "98e5762129afc4b198f6f1ea2894b760ee8328a227ad5481de5cbf8741717620", "generation_seconds_batch": 17.384654516004957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.268836498260498, "genuineness": 4.394378662109375, "identity": 0.5809300541877747, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/96de93739b7ac206aefa380fb8755dcb6b2d1612514d30d1a3c41e6051fbf010.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "97069d2679309a153a76b19d842bf5a595ed821506a961d2ee39dcbe3b7afac0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d24cd2e2733cad2ba9f05b7d871d8fcfc2aaa229fe5702582802a82a8016c5bd", "generation_seconds_batch": 16.661556991981342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3221020698547363, "target_z": 0.5156250863750386, "ce": 6.221180438995361, "genuineness": 4.557555198669434, "identity": 0.4933696389198303, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/97069d2679309a153a76b19d842bf5a595ed821506a961d2ee39dcbe3b7afac0.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "976ac2827e315ac68ac8f94f44b81a9d7318c38e306bd5c1d0b65f280ad774de", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b2e11a9a6f0080779aa52a150c96f5988a99ced5d4822e214220380517be1519", "generation_seconds_batch": 23.770009348052554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5159308910369873, "target_z": 0.6679687767177026, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.7614974975585938, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/976ac2827e315ac68ac8f94f44b81a9d7318c38e306bd5c1d0b65f280ad774de.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "976d8cce550f4e944aeff6f2b3e8dfe6e840222e02bede1c0f1235e2cebb4e4f", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1af9e668d056ad9de08c289c5620a901eb0b53ea703ecf8943679f9c90e3c63b", "generation_seconds_batch": 20.514942312031053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.867349624633789, "target_z": 0.1582031855723258, "ce": 6.117925643920898, "genuineness": 4.193545341491699, "identity": 0.3599889874458313, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/976d8cce550f4e944aeff6f2b3e8dfe6e840222e02bede1c0f1235e2cebb4e4f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "97a61f727da94c1eb02a5d5938a87caa7e705248d0d44b750d4d7ba89a457a08", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "374937daa357e1f9009a18dcb97555a4b3ac28874b29df186e1ecd483a04a466", "generation_seconds_batch": 20.371678151888773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.070269584655762, "genuineness": 4.595211505889893, "identity": 0.7604225277900696, "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_S_MONO_reg/97a61f727da94c1eb02a5d5938a87caa7e705248d0d44b750d4d7ba89a457a08.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "97fc0b4a372abf7bc87760a381e4bea83ff41170e16371c3dcafc45250d5a530", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1bb8c3e8f42f8919aeb7502658e6bb58bb0ac988ea5fd4dbe0be34ddf4fe627c", "generation_seconds_batch": 23.546955268946476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 5.847874641418457, "genuineness": 4.833701133728027, "identity": 0.6390575170516968, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/97fc0b4a372abf7bc87760a381e4bea83ff41170e16371c3dcafc45250d5a530.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "986264ab42a3df607367cd9bd74ea0f76d1662365542833eb41b2bbe44b67a27", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e6cfce5370938338fb3e6d04b242fcc24c3a8feda99000f9e87916d48fc9f599", "generation_seconds_batch": 21.929005658021197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.550720691680908, "target_z": 0.6953125256197148, "ce": 6.173524379730225, "genuineness": 4.18099308013916, "identity": 0.6786309480667114, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/986264ab42a3df607367cd9bd74ea0f76d1662365542833eb41b2bbe44b67a27.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "987b4d550100f70cf7d8f62e75eeca66272bdc64dcc35e51957437fee0643a2d", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5b1d615e1839a58b877cfea058878aba30b075be097e78aa5614c4203d7aaca7", "generation_seconds_batch": 17.649989911937155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.41180419921875, "genuineness": 4.344170093536377, "identity": 0.4392157196998596, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/987b4d550100f70cf7d8f62e75eeca66272bdc64dcc35e51957437fee0643a2d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "98828646dfab668cd8c9eef57a38cc04eb3148bcc5652c027c71025d5a399193", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b9a90c798dd31dde0a14b5752901b9ee472d329434d4252cb18716804a0b2d80", "generation_seconds_batch": 27.978949897922575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.316492557525635, "genuineness": 4.256305694580078, "identity": 0.36304548382759094, "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_S_MONO_reg/98828646dfab668cd8c9eef57a38cc04eb3148bcc5652c027c71025d5a399193.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "98f0a434222437455c93899e5a234265117dab4465f94d0c5709804905581ac0", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f920d02bc40b227abb45fb0f5e8eb348be78d0d958a67bb38508e1d8bab2045b", "generation_seconds_batch": 16.724924670998007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2624623775482178, "target_z": 0.46875006148731563, "ce": 6.356205940246582, "genuineness": 4.444586753845215, "identity": 0.3649752736091614, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/98f0a434222437455c93899e5a234265117dab4465f94d0c5709804905581ac0.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "98fde65a2931c00532aee73290ce57fee8c6b7ee31a3e10d055c57f305461512", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d471892a106cfeeaa0423838a3d303636902d986849635df33d76ff8f0ac5358", "generation_seconds_batch": 17.183784766006283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.070269584655762, "genuineness": 4.268857479095459, "identity": 0.7812620401382446, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/98fde65a2931c00532aee73290ce57fee8c6b7ee31a3e10d055c57f305461512.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "991e520b4a877aa570f4a957bf4471848f3ca4202be4f003ca449ca2f6bf2883", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b1ed36b493408def7a90df4a26460861da8578f86913697466026a0d74258a88", "generation_seconds_batch": 23.362556381034665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 5.696964263916016, "genuineness": 4.444586753845215, "identity": 0.38906413316726685, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/991e520b4a877aa570f4a957bf4471848f3ca4202be4f003ca449ca2f6bf2883.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "996f6c85a1571af6a2f4118d3727d7c0e0cad2ef9911063033ebf51ce1ab9fd5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d634e5742d3a731c37e024f2929ccdcd95a7da1b90b15194a2347c5363c2ea0b", "generation_seconds_batch": 17.022644135984592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4513211250305176, "target_z": 0.6171874216768718, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.5959051847457886, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/996f6c85a1571af6a2f4118d3727d7c0e0cad2ef9911063033ebf51ce1ab9fd5.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "c6cfcae5925fecee10990368e4ceb8e088076d43fa87fd6148e3020f8e963aa2", "generation_seconds_batch": 21.142702881013975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9269890785217285, "target_z": 0.2050780230701345, "ce": 6.332377910614014, "genuineness": 4.645420074462891, "identity": 0.6478621959686279, "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_S_MONO_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "9aa5fef3c904b601afa1b14157f1d1ea7f4e126ecc7f16a0dedcfba7cbdd6519", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "87cefffae3f77c5317f169436edd039ada02fda265deaa71864e13251349d9c0", "generation_seconds_batch": 17.39513406693004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.889714479446411, "target_z": 0.17578129648148264, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.35961246490478516, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9aa5fef3c904b601afa1b14157f1d1ea7f4e126ecc7f16a0dedcfba7cbdd6519.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "9ab41ebe1c540086cf1869df3e799bf8f27c6452290a053ed5ab17b3bde31352", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c54e6cf6f238ab1fa4c87fda45baa2f4ea3ccb9ffe569ee8310558f5919aea67", "generation_seconds_batch": 20.23984965996351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1471171379089355, "target_z": 1.1640623997171162, "ce": 6.300607204437256, "genuineness": 4.821148872375488, "identity": 0.679154634475708, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9ab41ebe1c540086cf1869df3e799bf8f27c6452290a053ed5ab17b3bde31352.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "9ae7ced8a49af4ef2b1a18851e4aa56058459484ae96f7092c2f7d9a6fabd552", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2177db7693446607725f5fb6000a1cc50e748bfb286cecbad15d7d4bab209323", "generation_seconds_batch": 24.065826344070956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2177324295043945, "target_z": 0.4335936522790877, "ce": 5.585766792297363, "genuineness": 4.268857479095459, "identity": 0.4588790237903595, "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_S_MONO_reg/9ae7ced8a49af4ef2b1a18851e4aa56058459484ae96f7092c2f7d9a6fabd552.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "9b0e7353ecb0d6d463e8cf342ef474247997c06ddb1c87e1c9e836139f3c8b37", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "53f1aa704869cc4889ae5af2b85e8044d4d3dcc4801533165d3f67f45973b40c", "generation_seconds_batch": 24.939828817965463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.554655075073242, "target_z": 1.4843751010148756, "ce": 5.863759994506836, "genuineness": 4.432034492492676, "identity": 0.8050237894058228, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9b0e7353ecb0d6d463e8cf342ef474247997c06ddb1c87e1c9e836139f3c8b37.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "9b91ab6646945521da585a8a31fc8fb041320095dc93da93e499ffc13c781976", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "298b6287952352c1dfa40ca52a265d04881493c4c98f4b9c33faa95e66926be4", "generation_seconds_batch": 21.893815565039404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.276336669921875, "target_z": 1.2656251097987778, "ce": 5.998785495758057, "genuineness": 4.758388519287109, "identity": 0.6603058576583862, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9b91ab6646945521da585a8a31fc8fb041320095dc93da93e499ffc13c781976.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "9ba938536e63d6e5673a925c41b421fc2d52ca7db6693fb8b1b65be12ad3b566", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "582a3e63e8dfc3e3ba963462a1ca9402ac4f438440ec74466134189fe94b464d", "generation_seconds_batch": 23.67651411902625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.1020402908325195, "genuineness": 4.494794845581055, "identity": 0.5326799154281616, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9ba938536e63d6e5673a925c41b421fc2d52ca7db6693fb8b1b65be12ad3b566.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "9bf4301a420a44937aee2c921831a40cd1f9f582450002d44b0c055c16c4b54c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "674f020ae8a01db21e647dc6ba7edfedcad7b4ffdbd1ff8142c4d7c226cd6c11", "generation_seconds_batch": 16.938465461018495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.815164804458618, "target_z": 0.11718746852435033, "ce": 6.09409761428833, "genuineness": 4.105680465698242, "identity": 0.7824233770370483, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9bf4301a420a44937aee2c921831a40cd1f9f582450002d44b0c055c16c4b54c.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "67eedf66501f91e796adcdceb7e172e70c418980441ffda060a7765d1ac4f2f2", "generation_seconds_batch": 16.48467895796057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3330776691436768, "target_z": -0.2617186859507129, "ce": 6.133810997009277, "genuineness": 4.469690799713135, "identity": 0.7044273018836975, "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_S_MONO_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "7ffd46dfe4e56e57b8f566e0aca1276a82392bf005084f3bc2c5090b5d7b08f8", "generation_seconds_batch": 16.678481868933886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 6.0305562019348145, "genuineness": 4.118232727050781, "identity": 0.6258512735366821, "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_S_MONO_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "9d1a982671262ef2246dbc9a9fe1a5f8579e05b9249338929667dcb4bf63b915", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d1aece406fd6076844c067f3e27641c1c3d56a49403ee78218bc65810b56e8f5", "generation_seconds_batch": 16.988274784060195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.312161922454834, "target_z": 0.5078124260688229, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.5083069205284119, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9d1a982671262ef2246dbc9a9fe1a5f8579e05b9249338929667dcb4bf63b915.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "9d347ac058f50dc9f64f03f1859687ee42eb787b5b783db46863345abd48c07e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "119de5acb1a01cdafed7d2ce3dac77e8a568853cca191276a1877164696f9737", "generation_seconds_batch": 16.93806311499793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.348263263702393, "genuineness": 4.7081804275512695, "identity": 0.725295901298523, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9d347ac058f50dc9f64f03f1859687ee42eb787b5b783db46863345abd48c07e.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "9d512de436f6d3c25f0e72b971e0b7f08385ed7de288c17d887d6406791a73f2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "085fa5ebd5a3cf4e3277d3f5dff9f2184f6060064a34655f6b6f0293133ce220", "generation_seconds_batch": 25.39391974499449, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4115614891052246, "target_z": 0.5859375300116659, "ce": 6.0146708488464355, "genuineness": 3.9927117824554443, "identity": 0.3356158435344696, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9d512de436f6d3c25f0e72b971e0b7f08385ed7de288c17d887d6406791a73f2.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "9e06e890f2491d8ad7ab7316e6f67e8368c313528e75f9f65ce5cac97759d0c4", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5c19cada028fbb50dc273d4782ff8057b810c31d216740cf21a84a80c95306f", "generation_seconds_batch": 27.677831275970675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.574534893035889, "target_z": 1.5000000468474786, "ce": 5.824047088623047, "genuineness": 4.406930446624756, "identity": 0.6988937258720398, "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_S_MONO_reg/9e06e890f2491d8ad7ab7316e6f67e8368c313528e75f9f65ce5cac97759d0c4.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "9e44d926a91891fb1e52f1fb8138e631763f10544bf657e1168c0e8a6a3a7aaf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ef68bb3b138965c589dd1c465a5f3276e4198cf6f215dcc7bfde62fbbda204e7", "generation_seconds_batch": 24.939828817965463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.465195655822754, "target_z": 1.4140626573782482, "ce": 6.252951145172119, "genuineness": 4.645420074462891, "identity": 0.7952386140823364, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9e44d926a91891fb1e52f1fb8138e631763f10544bf657e1168c0e8a6a3a7aaf.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "9eba5001cc798098049cafbf7bc1abcc52c2cad9ab82f023ef9f35178685f7ee", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ba58ba4fea38692b78230294a2b947f154cba59a49259731638f64008f68405e", "generation_seconds_batch": 18.44803942902945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 6.205295085906982, "genuineness": 4.143336772918701, "identity": 0.6775119304656982, "hypothesis": "La porte 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_S_MONO_reg/9eba5001cc798098049cafbf7bc1abcc52c2cad9ab82f023ef9f35178685f7ee.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "9ec669901ce5d73d87f393107ccbf91c572bdde80388d3e00d30236715939756", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4c0c503bd36d051bf5f2768568c33bbeec9cab37ee42c3717887afd52ad4677a", "generation_seconds_batch": 15.670681391027756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9667489528656006, "target_z": 0.2363281021252546, "ce": 6.3641486167907715, "genuineness": 4.444586753845215, "identity": 0.32265663146972656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9ec669901ce5d73d87f393107ccbf91c572bdde80388d3e00d30236715939756.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "9ee8d5f0fc0ed77c15aca524bc0a609321fe351f1d8d5717fc825dae96c1788a", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cb18b8c4d91cd874bd4e2c8d97acdf341ebf3e30129e264a149f47357f136c1", "generation_seconds_batch": 16.780870794085786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 6.427689552307129, "genuineness": 4.293961524963379, "identity": 0.4323023855686188, "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_S_MONO_reg/9ee8d5f0fc0ed77c15aca524bc0a609321fe351f1d8d5717fc825dae96c1788a.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "9eee2bcc4aeb08db85b0e96cebe95914b66fdc638fe7e2a9a3897b36813c5593", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "169ee3bbf553f049774e4ffbb091f148a4d2ec81fd294eb48083f5ae12817dce", "generation_seconds_batch": 26.92408249399159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.814129114151001, "target_z": 0.9023437137664033, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.7749353647232056, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9eee2bcc4aeb08db85b0e96cebe95914b66fdc638fe7e2a9a3897b36813c5593.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "9f130515b1749136e2bc858a06c1cb23de9fbc7d6893f63298cd3b1044167427", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a0062909e8c1ad312eec5e97e517c0d4a7bea31732ba83921c7897c258f1f6c2", "generation_seconds_batch": 22.009716066066176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.6759117841720581, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9f130515b1749136e2bc858a06c1cb23de9fbc7d6893f63298cd3b1044167427.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "9f2734588f558db49ff1283f58003d95f0defac4dfec855cb62cb3c9a5249f94", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "224e18af2e07634934b16ad9c1d3c7156d97f623270bf1cafe081f7dbb64d179", "generation_seconds_batch": 17.788812594022602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3022220134735107, "target_z": 0.49999995315252144, "ce": 6.427689552307129, "genuineness": 4.356721878051758, "identity": 0.32092010974884033, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9f2734588f558db49ff1283f58003d95f0defac4dfec855cb62cb3c9a5249f94.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "9f3d2329b7fce369a15ff60efd20fa7a9069fa51efe25f03ce3c5fdaec4ebd1a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f7ec2d184e8bba9a8c8f8c8cef94a52b086cde10a46881196946c7c013d4a115", "generation_seconds_batch": 18.007075214991346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.346951961517334, "target_z": 0.5351563623607494, "ce": 5.982900142669678, "genuineness": 4.017816066741943, "identity": 0.5973701477050781, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/9f3d2329b7fce369a15ff60efd20fa7a9069fa51efe25f03ce3c5fdaec4ebd1a.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "9f844d3c5ddf825bf3290e682873d99711ce78f54d14ef49b4f5ebd76b580dc2", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f77659088a7ce7dac5e5553f384cab846cc4731db55b97950d7e76f765086e3", "generation_seconds_batch": 27.68967806000728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.086154937744141, "genuineness": 4.582659721374512, "identity": 0.7648828029632568, "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_S_MONO_reg/9f844d3c5ddf825bf3290e682873d99711ce78f54d14ef49b4f5ebd76b580dc2.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "a0919a0ea04ccc2f68534ded6fc6a76b8f087d3fe7e2fdb2dd1a97992649507d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7e6b72132513b8ce67f9bd6f1723857340a77c88ff4db9a2023ff1840eaf4475", "generation_seconds_batch": 22.60140514408704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.77933931350708, "target_z": 0.874999964864391, "ce": 6.260893821716309, "genuineness": 4.325342178344727, "identity": 0.37532684206962585, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a0919a0ea04ccc2f68534ded6fc6a76b8f087d3fe7e2fdb2dd1a97992649507d.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a0c0b5d5d039ac3e4d97876d486e0fff3e2e44ea188b0ceae843785abf06318d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f328e4687c377349e00a0335d3750a25afe7bd37c18983517165bab2ebc1ab5d", "generation_seconds_batch": 23.50982217793353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4513211250305176, "target_z": 0.6171874216768718, "ce": 6.24500846862793, "genuineness": 4.545003414154053, "identity": 0.5039471387863159, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a0c0b5d5d039ac3e4d97876d486e0fff3e2e44ea188b0ceae843785abf06318d.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "a0cafc4a388e4b6e5ac798e50306786f40c6e0e6ae93a473205cee7ed22f530c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3b3422ada228231b9f9e625696925df80fc9a230683697a77cfe138bbc1539b1", "generation_seconds_batch": 16.809002202004194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.118333101272583, "target_z": 0.3554687357261589, "ce": 6.221180438995361, "genuineness": 4.124508857727051, "identity": 0.5410361289978027, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a0cafc4a388e4b6e5ac798e50306786f40c6e0e6ae93a473205cee7ed22f530c.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "a118bdd8ef5748bc540f7e8fa6986aa7de14887117dd2f20cb742b0fc4f1b382", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "94a4d2a339a59c35c940ba14f060d72bb171b165010adcbb33936aff06087887", "generation_seconds_batch": 26.787865124060772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 6.300607204437256, "genuineness": 4.607763767242432, "identity": 0.6051819324493408, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a118bdd8ef5748bc540f7e8fa6986aa7de14887117dd2f20cb742b0fc4f1b382.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a13c395d7fc61aef106763330e2c6cf9f12a00da40d5dfdb1bb9ce062f2b66ed", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8398de8722f4375bc067fa3f14685165dc6cd2b6705554f138baca88f01e35cb", "generation_seconds_batch": 25.11648467509076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2962164878845215, "target_z": 1.2812500556313808, "ce": 5.506340026855469, "genuineness": 4.18099308013916, "identity": 0.7012244462966919, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a13c395d7fc61aef106763330e2c6cf9f12a00da40d5dfdb1bb9ce062f2b66ed.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "a1523addb05e95c8689b4ae5deae87f6001d1fe96c64c51285216fb90cbf21af", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "272e4d4720ba2684aa7f87801ec0d086a335552a251f636c3b9706135cccea2d", "generation_seconds_batch": 27.919004468014464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 5.669164657592773, "genuineness": 4.281409740447998, "identity": 0.6225255131721497, "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_S_MONO_reg/a1523addb05e95c8689b4ae5deae87f6001d1fe96c64c51285216fb90cbf21af.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "a1d1e562edd1453d061ef6858bbd4e63b262a40481e9eb7f5cfdaab8158f1bfd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6104a02f61883bbd70e3445478089fff3ad10eda6944fd1dc65d200e4e0b37b4", "generation_seconds_batch": 23.253210914088413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.6320681571960449, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a1d1e562edd1453d061ef6858bbd4e63b262a40481e9eb7f5cfdaab8158f1bfd.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "a1f9371f66ebdb6ea0cf58edcfd141230a31cf7d45b53d3a4cabdf0aa8bb9655", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "79f9e6af9c2c909da160fcb8527d4cbae58703ca171e3b683f14185f761600d7", "generation_seconds_batch": 21.78256533597596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.157057285308838, "target_z": 1.1718750600233319, "ce": 5.8875885009765625, "genuineness": 4.143336772918701, "identity": 0.38958999514579773, "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_S_MONO_reg/a1f9371f66ebdb6ea0cf58edcfd141230a31cf7d45b53d3a4cabdf0aa8bb9655.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "a236b54e45de010f14d2f1147319f2d60127a873d4bdcfb3c5c400b1825df652", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bca3550f523ea33916cc7060658cea85379d485ce48f3000e096f9f2dc5ea5d4", "generation_seconds_batch": 16.93806311499793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345916271209717, "target_z": 1.3203126076028024, "ce": 6.324435234069824, "genuineness": 4.7332844734191895, "identity": 0.7629618048667908, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a236b54e45de010f14d2f1147319f2d60127a873d4bdcfb3c5c400b1825df652.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "a29cf439615ac3d33eb6a97eea9d04e22bdb9678e289afd8d31d14cbad6ebe85", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4a57ac7735f58f462f4d766f6c82e1daeda1c116a3673584c16fcb0fea9965ff", "generation_seconds_batch": 24.133278941968456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.524835109710693, "target_z": 1.460937494876057, "ce": 6.221180438995361, "genuineness": 4.15588903427124, "identity": 0.4125952422618866, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are", "wer": 0.05555555555555555, "duration_error_s": 0.120000000000001, "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_S_MONO_reg/a29cf439615ac3d33eb6a97eea9d04e22bdb9678e289afd8d31d14cbad6ebe85.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "a3313b197411c18658e538545fc5a9374a0e9c0666d21321978b557855ed7c0b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "73030566cb6211ec3df3cfb8e36fb6761bc2961aca01e7a41227591f4ea76cef", "generation_seconds_batch": 24.753312492044643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.77933931350708, "target_z": 0.874999964864391, "ce": 6.109982967376709, "genuineness": 4.6956281661987305, "identity": 0.4964280128479004, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a3313b197411c18658e538545fc5a9374a0e9c0666d21321978b557855ed7c0b.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "a3e29a5a94d231b2cad58b28268da4f8aece1a07f95628ce4fbddc653974c838", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c833cc66b4e6c991058eebd5b1c6d4d1ac138c12065e725679158f9c2a48f812", "generation_seconds_batch": 19.73513227608055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8748044967651367, "target_z": 0.16406249341207332, "ce": 6.046441555023193, "genuineness": 4.7081804275512695, "identity": 0.6159617900848389, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a3e29a5a94d231b2cad58b28268da4f8aece1a07f95628ce4fbddc653974c838.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a414eeee879cafa3472bb7f7989caeb25c6702103f5abcb07d0938796ba05fff", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f954911ebe0a5ce424ce562b8926b881be7c9160252f07e08d8df7b95f909227", "generation_seconds_batch": 30.90115448506549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 5.919358730316162, "genuineness": 4.469690799713135, "identity": 0.7649141550064087, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a414eeee879cafa3472bb7f7989caeb25c6702103f5abcb07d0938796ba05fff.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "a41a512f26d9938afd4bbfa1198dc90141b7d5b5d8354125381f1a6143d66e74", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e1b88d60d6857de37c633e30533592ded8f07aa849cb34648660fc5c565ae47d", "generation_seconds_batch": 17.814961828989908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.227672576904297, "target_z": 0.4414063125853034, "ce": 6.117925643920898, "genuineness": 4.469690799713135, "identity": 0.6874761581420898, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a41a512f26d9938afd4bbfa1198dc90141b7d5b5d8354125381f1a6143d66e74.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "a42cf9e9682ffe9239b6a474b8b450014e79c5d12c2da2fee04d147587740152", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "faa96e1724ea1e7cfcd863bb7ec3109508121ac8c8ca256ad811874d35b830b0", "generation_seconds_batch": 22.848785149049945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.284721851348877, "genuineness": 4.413206577301025, "identity": 0.42091965675354004, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a42cf9e9682ffe9239b6a474b8b450014e79c5d12c2da2fee04d147587740152.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "a47f829a17875bfcf08f7b0c64cde47960705ad0032c1e076a0647df5eac889a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1e83b140973d3fbaaa2cf825454ff57b673f581694a4523b6e0a84e6741f8a77", "generation_seconds_batch": 17.292658876045607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.263498067855835, "target_z": -0.31640618375473734, "ce": 5.998785495758057, "genuineness": 4.15588903427124, "identity": 0.7019727826118469, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a47f829a17875bfcf08f7b0c64cde47960705ad0032c1e076a0647df5eac889a.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "a4c560fa555302649527dcdbabb73ecf16b208fd7fb3758e39f1ee7760617727", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0982affc4013e527de43ebadfc15d93e1b6ae3840f71fbe412d88912cebd106c", "generation_seconds_batch": 25.185939830960706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.6102290153503418, "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_S_MONO_reg/a4c560fa555302649527dcdbabb73ecf16b208fd7fb3758e39f1ee7760617727.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "a4cd78e5b612c2da7207ee6a29ffeafb2a0acb0d5a339602848dffe2ea27453d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d552e5bf318d73166e04af7a037601f89332f58fe057ab009eb123df33f001e1", "generation_seconds_batch": 21.52669397299178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6352100372314453, "target_z": 0.7617186391032343, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.6264582872390747, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a4cd78e5b612c2da7207ee6a29ffeafb2a0acb0d5a339602848dffe2ea27453d.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "a51443f32034894a6c08978c83b3c31bc5a9abdc8e3d165abb1c50a3e191a09d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "707bd7147cd94067465be2520c47c0b6bf964ea098f7fad882032f86fa54b8c2", "generation_seconds_batch": 18.12347157509066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.5670641660690308, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a51443f32034894a6c08978c83b3c31bc5a9abdc8e3d165abb1c50a3e191a09d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "a58618323863a21754edd89526c2734c04141210f4a2b9966979410e579dbbe4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "52d05f4a56a5ca5fb2ae3e7d8981dd5a222886f8042f3b3e068d65158afb9969", "generation_seconds_batch": 25.365305499988608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.809159278869629, "target_z": 0.8984375710032098, "ce": 6.0305562019348145, "genuineness": 4.457139015197754, "identity": 0.2581251263618469, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a58618323863a21754edd89526c2734c04141210f4a2b9966979410e579dbbe4.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "a589f463d8d7e528403cfd6a365f22bfccc14062bde3eeb65c94d9aac5064086", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a67ab010c3c2895ade3094f8305910e979fcfcb7a52863653beef5b32a9a2c25", "generation_seconds_batch": 23.681950527010486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.923468589782715, "target_z": 0.9882812906255478, "ce": 5.967014789581299, "genuineness": 4.111956596374512, "identity": 0.7138084769248962, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a589f463d8d7e528403cfd6a365f22bfccc14062bde3eeb65c94d9aac5064086.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "a5eff6f6c4556dc8cef3925a503ad8962cf8a4e62508b39b56c415aff3844ce0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9008f1a1a4235903cf0c963250b2f38cb631eede5f10df8a3d18ea51c8583edd", "generation_seconds_batch": 22.532765422016382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.888678789138794, "target_z": 0.9609375417235356, "ce": 6.252951145172119, "genuineness": 4.1998209953308105, "identity": 0.661597490310669, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a5eff6f6c4556dc8cef3925a503ad8962cf8a4e62508b39b56c415aff3844ce0.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "a64c4b248918b898451c471536e642a1b4f9c9ba2b487f744db8bdf48562cd68", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5a061dc9e16f2f765caf43aae8476f2fac9735f8efc114d2805e9e3f233d88d5", "generation_seconds_batch": 23.92005821201019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.6624517440795898, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a64c4b248918b898451c471536e642a1b4f9c9ba2b487f744db8bdf48562cd68.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "a77bb8b2fa352be88dbbb7e758550c2e6c1c12837bdedfab5cf2cb8c0774a789", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b31dfbb72c19ad8c06f36f5e71522192d9fd66044a8dedb326e6d3fdbd62ea8b", "generation_seconds_batch": 24.103951104916632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.23706579208374, "genuineness": 4.281409740447998, "identity": 0.1191810742020607, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a77bb8b2fa352be88dbbb7e758550c2e6c1c12837bdedfab5cf2cb8c0774a789.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a85bda73d431cb38044a297edb4d2f6942fdaeb667912d6ca503507f4ca8f359", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bd0bfbda96713f7f0e43760c28773dca5baaaabf4ca1e6bda46991ed23e981eb", "generation_seconds_batch": 17.693446008022875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.7139690518379211, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a85bda73d431cb38044a297edb4d2f6942fdaeb667912d6ca503507f4ca8f359.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "a898ff6012475ccc914717bc701bd98b2fc5c0200a9244fa0221d7e67c9b9e5a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cc472e2ff28d19970269ac3a047d87910a402cd45d891e87523b9bb98309f242", "generation_seconds_batch": 21.29275748692453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.268836498260498, "genuineness": 4.306513786315918, "identity": 0.24798645079135895, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a898ff6012475ccc914717bc701bd98b2fc5c0200a9244fa0221d7e67c9b9e5a.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "a8a4b02d80feb877aca63c497c4f2b0c5b27f029f3596e5dc35b83618c224435", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bfbf5d4ec96a6f303c6101bd40afd84b343b08bc8181978da7293ec97b4baee", "generation_seconds_batch": 27.634806237067096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.594414710998535, "target_z": 1.5156249926800816, "ce": 5.7684478759765625, "genuineness": 4.796044826507568, "identity": 0.6924469470977783, "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_S_MONO_reg/a8a4b02d80feb877aca63c497c4f2b0c5b27f029f3596e5dc35b83618c224435.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "a8af9214cbca37829072cdeaf2011819929c38cfa86633a0e5a1de8d7eef8edc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7b31e38626e40a97da84096d95b60d9761d2bb3779bbfdab133f8afa2bc1d436", "generation_seconds_batch": 24.745266709011048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2077927589416504, "target_z": 0.4257813667527005, "ce": 6.062326908111572, "genuineness": 4.2751336097717285, "identity": 0.7532327175140381, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a8af9214cbca37829072cdeaf2011819929c38cfa86633a0e5a1de8d7eef8edc.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "5efc92ca947ff7e05a350ade57f6e29d5f77e352ff9575194bb618a01b6b94ca", "generation_seconds_batch": 16.563369904994033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9742040634155273, "target_z": 0.2421875973549164, "ce": 6.086154937744141, "genuineness": 4.2249250411987305, "identity": 0.7887007594108582, "hypothesis": "Ik schreef het nummer op de aanvelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a8fd823d664ad1742c4535048787b690471938c9c48a1ecdc1eaa3dd070c1868", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bc24fb9440dce2b2c01d6b1908b93fc601567cedaa9fc982556bb95624edd1e3", "generation_seconds_batch": 30.90115448506549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.346951961517334, "target_z": 0.5351563623607494, "ce": 6.173524379730225, "genuineness": 4.607763767242432, "identity": 0.5086824297904968, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a8fd823d664ad1742c4535048787b690471938c9c48a1ecdc1eaa3dd070c1868.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "1c24d931c228d60196e66479dce8e8e782a2fe1070ad0915d707bb36c371fe76", "generation_seconds_batch": 15.167469165055081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7797536849975586, "target_z": 0.08935538150572106, "ce": 5.982900142669678, "genuineness": 4.482243061065674, "identity": 0.7575641870498657, "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_S_MONO_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "a95a72361ddd988a7917afef3c05d65422b11dd2aa2362e7f788327f4df2adaf", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dd8d83418a093e198987d3d124f92ea2511ae80aa94cd074dc05e5047bcd3998", "generation_seconds_batch": 17.703343112021685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.312161922454834, "target_z": 0.5078124260688229, "ce": 5.871703147888184, "genuineness": 4.074300289154053, "identity": 0.4715631604194641, "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_S_MONO_reg/a95a72361ddd988a7917afef3c05d65422b11dd2aa2362e7f788327f4df2adaf.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a96345e403f246bdf099fd97df70e801ab7977384ee24ef7c0adf427423814a0", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4eb0c10b5b39d5bd02fb16d482e08c32854cb15f6a9ed69cb00c175bfe74eda6", "generation_seconds_batch": 16.285176000907086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9046244621276855, "target_z": 0.18750009955089197, "ce": 6.284721851348877, "genuineness": 4.545003414154053, "identity": 0.6845449805259705, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a96345e403f246bdf099fd97df70e801ab7977384ee24ef7c0adf427423814a0.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "a9a4a5421792072b69e43e3bd25128ba9dfa4d6e942c6b7358573a1dbffdcf01", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "2f68d19286fa9b4799a74349e47663011594e5da291604dad806191508614031", "generation_seconds_batch": 23.160940138972364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.133810997009277, "genuineness": 4.400654315948486, "identity": 0.6609483957290649, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a9a4a5421792072b69e43e3bd25128ba9dfa4d6e942c6b7358573a1dbffdcf01.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "a9c71e8a175319c157c2308f761b2d83dd5aa1e866bdd4c9c91fc94fa2613507", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b63c56ec1bca400cae2e5b3cc96f2c63a4f8778cb49a2ea00df64c837279d85d", "generation_seconds_batch": 18.01490982202813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9184985160827637, "target_z": 0.98437496047244, "ce": 6.09409761428833, "genuineness": 4.168440818786621, "identity": 0.7654982805252075, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a9c71e8a175319c157c2308f761b2d83dd5aa1e866bdd4c9c91fc94fa2613507.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "a9ed3cb9e8e9bbae7ed5e7928d7a66cefeb897e558acc40d2e0cd4c4f7b4f3b4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a3af5600c5f6aa44f1e6929895887f11ff73eb6ae7a2a785bd90946963b21375", "generation_seconds_batch": 23.137947648065165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.7850065231323242, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/a9ed3cb9e8e9bbae7ed5e7928d7a66cefeb897e558acc40d2e0cd4c4f7b4f3b4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "aa5c48e62ca1eebee847fbe1e999d81017b7f0e0230998e7924a6e509edc79f7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "be364996660e10f5fceaeddc6485a8e60230faaf510811364ea522f0d8e32fcb", "generation_seconds_batch": 17.594369477941655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.443574905395508, "genuineness": 4.570107460021973, "identity": 0.5498344898223877, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/aa5c48e62ca1eebee847fbe1e999d81017b7f0e0230998e7924a6e509edc79f7.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "aaa19c69f56f1d8fb2847a7230f706f545cefc73a510b83442d4224209152da4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "55f694f7c89babb654016b2d2859caed1227fff2018928b7c193ef8e3bc05530", "generation_seconds_batch": 27.424577189958654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1034231185913086, "target_z": 0.3437499326567496, "ce": 6.181467056274414, "genuineness": 4.582659721374512, "identity": 0.5562306046485901, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/aaa19c69f56f1d8fb2847a7230f706f545cefc73a510b83442d4224209152da4.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "ab548e58a3f9cbe99aa34f619a24b215f9689e06c2a4ceee6dea7b260416eb21", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cf50f1d38fe0462ce5effb2bf7cc0a0725e902834ff4cede42f2d633edda2a0a", "generation_seconds_batch": 16.62805140390992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.141753673553467, "genuineness": 4.319066047668457, "identity": 0.6608524322509766, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ab548e58a3f9cbe99aa34f619a24b215f9689e06c2a4ceee6dea7b260416eb21.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "ab566d46c480403d457674a2e2c3942c6c34e89e364aad157394087e4ec448e2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b53696457480086d02425cd2caf1687359556efde2b4bf8ab00b996305b5c97e", "generation_seconds_batch": 26.98538244701922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 6.340320587158203, "genuineness": 4.337893962860107, "identity": 0.3484572172164917, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ab566d46c480403d457674a2e2c3942c6c34e89e364aad157394087e4ec448e2.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "f1806f39a9ba4ebf24f8049c148a034d0b5d880632d4635185bed016e9d40275", "generation_seconds_batch": 18.197057637036778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1779727935791016, "target_z": 0.4023437606138819, "ce": 6.41180419921875, "genuineness": 4.15588903427124, "identity": 0.4536025822162628, "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_S_MONO_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "abc984faf1edd22ae09b6ba1c3da65031187e0a85c0e4ce0ebf7f220f82058b7", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8f16c84a0d7b75739c9daed40e8eecbd1e1e6d3cdce9db72dac9ae2e585a795d", "generation_seconds_batch": 16.875278149032965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0537235736846924, "target_z": 0.3046875680752423, "ce": 6.260893821716309, "genuineness": 4.068024158477783, "identity": 0.3980521857738495, "hypothesis": "The 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_S_MONO_reg/abc984faf1edd22ae09b6ba1c3da65031187e0a85c0e4ce0ebf7f220f82058b7.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "ac173b9a6d545d96cbfa3cc5bb06c451bbecc2cc78991e0bf00d1078e57a0bea", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "614451b3767b09ecaa3738795cf7c58fd2954bacbef21b4b1889284cacf23188", "generation_seconds_batch": 17.21253404498566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4950151443481445, "target_z": 1.4374998887372383, "ce": 5.77639102935791, "genuineness": 4.871357440948486, "identity": 0.652420163154602, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ac173b9a6d545d96cbfa3cc5bb06c451bbecc2cc78991e0bf00d1078e57a0bea.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "ac3b056d963a75d2229a7f7f76ed47b564199721ac7e39710ff4eb358ecd9341", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e62613c0719200e18c7490e18758b6c284fc30865fec9cd8170a8026712f8c65", "generation_seconds_batch": 27.399289103923365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.6441144943237305, "target_z": 1.554687544651503, "ce": 6.157639026641846, "genuineness": 4.15588903427124, "identity": 0.546103835105896, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ac3b056d963a75d2229a7f7f76ed47b564199721ac7e39710ff4eb358ecd9341.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "ac402b508b20f8f4ce93a5ef296fb23bb7bacaf5ad4df0ff44f991ade5bbab3c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c2ec5ff3f6eb87c3d76428afaedd5cec73010a5312983e5f7ae924dd1708bc50", "generation_seconds_batch": 27.298380355001427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4513211250305176, "target_z": 0.6171874216768718, "ce": 6.078212261199951, "genuineness": 4.218649387359619, "identity": 0.5131424069404602, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ac402b508b20f8f4ce93a5ef296fb23bb7bacaf5ad4df0ff44f991ade5bbab3c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "ac7fab6d5547a5f7bdd947a6c2208afeb7d882244e2540005a2162dd2511590c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "46487089923bb5fd515b4f45ee065830ea7269454db650493ec8acd5815e5253", "generation_seconds_batch": 21.15967216505669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 6.062326908111572, "genuineness": 4.457139015197754, "identity": 0.6266144514083862, "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_S_MONO_reg/ac7fab6d5547a5f7bdd947a6c2208afeb7d882244e2540005a2162dd2511590c.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "acca87748b9ef4524d2cbf7c1379d3e5105401baba8786a01a3cd61b5f5d2939", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b12fa7523ae0d07b43bb37c6ad06bac34363fec9789a4ebabe8389a8f992221a", "generation_seconds_batch": 26.92408249399159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 6.062326908111572, "genuineness": 4.331617832183838, "identity": 0.6679319143295288, "hypothesis": "La porte 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_S_MONO_reg/acca87748b9ef4524d2cbf7c1379d3e5105401baba8786a01a3cd61b5f5d2939.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "acf197ae22fab21341fd51f8e79005e35d2123aaf368916e21ea87e08f2f72d2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "06e3fa75e284564dad7d8f9e14b792abae5c18c41083fbf73ca2b8caa3548b04", "generation_seconds_batch": 16.14919034903869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.381741523742676, "target_z": 0.5624999238728473, "ce": 6.308549880981445, "genuineness": 3.8671910762786865, "identity": 0.23591357469558716, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/acf197ae22fab21341fd51f8e79005e35d2123aaf368916e21ea87e08f2f72d2.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "3ec015e9e8c365b76dacd950e6ecc1bae9f87310af68afcc0260e47689816054", "generation_seconds_batch": 20.380800051032566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.88474440574646, "target_z": 0.1718749663283748, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.6349900960922241, "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_S_MONO_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "ae23a5592207d94d879069ce972d18f20a6531a886be445cac4655e406871d3b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0d08e05570450c311ab1a9b46389e32ef4881f2aee13df9962a1ceb45b1d284a", "generation_seconds_batch": 16.230958738015033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.6686273813247681, "hypothesis": "En concreto la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ae23a5592207d94d879069ce972d18f20a6531a886be445cac4655e406871d3b.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "ec64425fba1021d840f470efac110f7a9cfd366d0275df0fd351fef123e68012", "generation_seconds_batch": 17.713009259081446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.026388645172119, "target_z": 0.2832031270129776, "ce": 6.403861999511719, "genuineness": 3.9550557136535645, "identity": 0.3610089421272278, "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_S_MONO_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "ae582e38cf41e06e05289a18e5ff829f5b466439515fbbb2f4b03cba3f879cc4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d0d8b004e20250c7861315c37b8ffd941a4d426fcf51748c8930b12106186c61", "generation_seconds_batch": 23.678326043998823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6103603839874268, "target_z": 0.7421875505074378, "ce": 6.221180438995361, "genuineness": 3.848362922668457, "identity": 0.45935213565826416, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ae582e38cf41e06e05289a18e5ff829f5b466439515fbbb2f4b03cba3f879cc4.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "aefa5967b7c28dbde208c6dc03550c2895d33511ae88f4ebccd3198a45d9de01", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "526e5793ede79fc351753e03bc412e4aa7ec0f459fac63df58bd02742119c2ad", "generation_seconds_batch": 17.333970076055266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.574534893035889, "target_z": 1.5000000468474786, "ce": 5.95112943649292, "genuineness": 4.607763767242432, "identity": 0.658571183681488, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/aefa5967b7c28dbde208c6dc03550c2895d33511ae88f4ebccd3198a45d9de01.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b045f0ea5e1235746d1db356f3b3a02383749d246d6da6b825cf43936b252c1d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "834afb17fe478668ac92a1382637e2bb96b881dcd69b151caa442a160ee0dcac", "generation_seconds_batch": 21.72323978599161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 5.80021858215332, "genuineness": 4.607763767242432, "identity": 0.832979679107666, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b045f0ea5e1235746d1db356f3b3a02383749d246d6da6b825cf43936b252c1d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "b0ca674e5e36b28adade53b7d52b15b7668f52b4a10f60aa92fe9e5b9d696d47", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1bb15d65b9d403f3f56ed9cd3f0547d9b41b3c3073a09ee4753e332c614d4de0", "generation_seconds_batch": 21.70670681493357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 5.824047088623047, "genuineness": 3.986435890197754, "identity": 0.6163896322250366, "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_S_MONO_reg/b0ca674e5e36b28adade53b7d52b15b7668f52b4a10f60aa92fe9e5b9d696d47.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "b115fc7d9125e5e2d7dc9d3fb52ec2bb6ef59879530a67009143c8fc1f14f7ef", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "be64c72e9804066b192474ca94c7f5b4cdb24986c054185eba9a0826491d8500", "generation_seconds_batch": 21.52669397299178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.6258143186569214, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b115fc7d9125e5e2d7dc9d3fb52ec2bb6ef59879530a67009143c8fc1f14f7ef.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "b1587b66d9c427743ee5aa58d40b25f71b5da72c35bfab79538cf497470920d1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a006d8b4ca6615a5da82bbb3e262d46a33dbbaecfb7c65d14c4d89a2debfa873", "generation_seconds_batch": 22.12607882800512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.455255508422852, "target_z": 1.4062499970720326, "ce": 5.601652145385742, "genuineness": 4.356721878051758, "identity": 0.6000596284866333, "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_S_MONO_reg/b1587b66d9c427743ee5aa58d40b25f71b5da72c35bfab79538cf497470920d1.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "b17bf4bc73fd8fc4b038977bf1e240c664954e6dd1fbdabe44723503e6c34552", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e631fdfc05b2686cac6290cdb5fb4d87ecb9c8a3c5db6cc01872103fbf65ccfe", "generation_seconds_batch": 17.107972488971427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 5.681078910827637, "genuineness": 4.086852550506592, "identity": 0.5137577056884766, "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_S_MONO_reg/b17bf4bc73fd8fc4b038977bf1e240c664954e6dd1fbdabe44723503e6c34552.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b1c25fa03926a75ed0064ed095e8564bafbd1e788426eb8cb5851ba93962e160", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7dad31c138712dfb73ef09a9b5be63a77effa06121e8d04c4ed1f159f169e4d0", "generation_seconds_batch": 20.06638449395541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 5.8875885009765625, "genuineness": 4.632867813110352, "identity": 0.6652784943580627, "hypothesis": "He encontrado la cajita. Estaba junto a a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b1c25fa03926a75ed0064ed095e8564bafbd1e788426eb8cb5851ba93962e160.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "b21d58d791b7237377e56e7d89351297de2ff9b4d66671f27ebeb51f9f055fbb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f607c8118c5b3796a63a8360c02fb1b633a0e3e6274a8a806a314ab9387671b5", "generation_seconds_batch": 17.326142848003656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.054384231567383, "genuineness": 4.13078498840332, "identity": 0.7470439672470093, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. We'll regard ensemble what it contiens.", "wer": 0.3157894736842105, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b21d58d791b7237377e56e7d89351297de2ff9b4d66671f27ebeb51f9f055fbb.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "b270adf9dd85f93cb214f57422f31d1e51dd52f8a956c26f42560f4dd62fe2db", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "81bd12f2830ccd2b8dc797ecf1dbc59104ebd70da3dde6f98a3958c02596e691", "generation_seconds_batch": 24.602987863007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4960508346557617, "target_z": 0.6523436434951855, "ce": 6.300607204437256, "genuineness": 4.344170093536377, "identity": 0.39562898874282837, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b270adf9dd85f93cb214f57422f31d1e51dd52f8a956c26f42560f4dd62fe2db.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "b2f8c324c1326109d4b7dd61d43e1c24250410ee9af99b8a889905d5388d2784", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "70f6dbf64be268bd3565a76064bb83d9b208106549d30e64908b774aea309763", "generation_seconds_batch": 16.96149707690347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.905866861343384, "target_z": 0.18847658839421177, "ce": 6.292664527893066, "genuineness": 4.582659721374512, "identity": 0.41476190090179443, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b2f8c324c1326109d4b7dd61d43e1c24250410ee9af99b8a889905d5388d2784.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "b32622494d57e383d7f897ad52ec614f94bbd5614a593f2803a7eea0b6fce083", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fe4fe08e24eb1fe4382b713a18799d7ebbdd321cd5d8082b3602fa9a17b2e579", "generation_seconds_batch": 26.754768899059854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197852611541748, "target_z": 0.41796870644648476, "ce": 6.2767791748046875, "genuineness": 4.362998008728027, "identity": 0.19921863079071045, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b32622494d57e383d7f897ad52ec614f94bbd5614a593f2803a7eea0b6fce083.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b346b0066366057dc48d5cd8d05ba395cb53ff46daf0fd66da454f1d25a890f1", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9cf1247ebcc0e96c06e3511ef4b786e9e68046c4fed8e369407c5956543e493f", "generation_seconds_batch": 15.092310701031238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.7012295722961426, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b346b0066366057dc48d5cd8d05ba395cb53ff46daf0fd66da454f1d25a890f1.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "b366e941cda3015defcf341c42a84efd91b42340970e1e687369a37d5adadba8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f06ef2171bc169e810a536fe74c35a0cf4d26164669d58a6c68ed41c57050255", "generation_seconds_batch": 25.345219173002988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1730027198791504, "target_z": 0.398437430460774, "ce": 6.23706579208374, "genuineness": 4.061748504638672, "identity": 0.39138683676719666, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b366e941cda3015defcf341c42a84efd91b42340970e1e687369a37d5adadba8.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "b40819913de012da0d280fce88a052e0628c2a2d6e84b68298a05008b587e50e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c2da441913447e2f760c13eb5d96bd9f6f206c3cfa23938e5ac88088799b99a6", "generation_seconds_batch": 22.34321841993369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.6542913317680359, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b40819913de012da0d280fce88a052e0628c2a2d6e84b68298a05008b587e50e.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "aa3139f230fe40924f83e09fbb6ca5e641130f498d6033932463ddbf9fd3e8f5", "generation_seconds_batch": 14.915647619985975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242582321166992, "target_z": 0.45312492826479844, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.4981945753097534, "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_S_MONO_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "b498c161ac4365ac5d6fa497dd87a3ae1a6f279f0d2f26510370be092816a0bc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e8754f65b140e4cd3de500bd747e4a611be00efd17332b8894851a283ae01eae", "generation_seconds_batch": 24.447705950937234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.306156158447266, "target_z": 1.289062341157768, "ce": 5.601652145385742, "genuineness": 4.256305694580078, "identity": 0.4344947636127472, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b498c161ac4365ac5d6fa497dd87a3ae1a6f279f0d2f26510370be092816a0bc.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "b5311f344dd3006acc3ef908816843bbf4a8aa632e244282824a32e1194094dd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "921f1b3e6e8503fe5907bf916ec858ad36c404f38814485c38792eca71cd5151", "generation_seconds_batch": 17.15684340405278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.268836498260498, "genuineness": 4.193545341491699, "identity": 0.7057708501815796, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b5311f344dd3006acc3ef908816843bbf4a8aa632e244282824a32e1194094dd.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b5ca21545b06a99e6b269e49b01232c0d14a8fc911ddb61c124778b81efbd780", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d4f70042b93647f86f1cdbe30182848149e3d15334015448ff0a056a8a3d106f", "generation_seconds_batch": 24.48224454803858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.24500846862793, "genuineness": 4.532451152801514, "identity": 0.6255266666412354, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b5ca21545b06a99e6b269e49b01232c0d14a8fc911ddb61c124778b81efbd780.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "b5e4be0bda42cdeea527220aeb89bccf5c20c465863c1a44cd0c30474f92c808", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dc2a538bf13a88b9d957060d33c4878c975a1b996b72a6682488cd736bf38e45", "generation_seconds_batch": 16.706398878013715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2823421955108643, "target_z": 0.4843750073199185, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.7277534604072571, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b5e4be0bda42cdeea527220aeb89bccf5c20c465863c1a44cd0c30474f92c808.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "b5ea24d7eb0ae1adefe3bb6a6ec0fab38cd2ffe236d148a0b8ade9cad71da467", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c40774ffa4de901224ef2a18eeba06f23017ee1a7811f949ded7ca18a99026f7", "generation_seconds_batch": 18.007075214991346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.417567014694214, "target_z": -0.1953125724671934, "ce": 5.855817794799805, "genuineness": 4.143336772918701, "identity": 0.6522631049156189, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b5ea24d7eb0ae1adefe3bb6a6ec0fab38cd2ffe236d148a0b8ade9cad71da467.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "b5ed8de8a97a299750cfea9e48cd7023a2534a253e4ed2d01dc220fda945a646", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0f678d829d66c36cd544edaddaa1684b6d64939052d1700ad0e735cc05c8808b", "generation_seconds_batch": 20.51666610792745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7061359882354736, "target_z": 0.031494060723514336, "ce": 6.229123115539551, "genuineness": 4.3127899169921875, "identity": 0.7404721975326538, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b5ed8de8a97a299750cfea9e48cd7023a2534a253e4ed2d01dc220fda945a646.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "b65d4448ef379c134050a2c2c878308f5d1fa5cc1ce5023a00c782e36d54074a", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccc869cd9191423550186ae615bfbc0057325e102cd8b739bb179aa740422e4d", "generation_seconds_batch": 27.77285269391723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.316492557525635, "genuineness": 4.7081804275512695, "identity": 0.7124035954475403, "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_S_MONO_reg/b65d4448ef379c134050a2c2c878308f5d1fa5cc1ce5023a00c782e36d54074a.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b6a8208322ebe5771e7148bb74a448bebb31c38c254721eced59019b84f97075", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "20dec81bb457ddb8dca747e3402842e7722891a1dc9391d7eb088a1e1565a2f6", "generation_seconds_batch": 20.742347405059263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.324435234069824, "genuineness": 4.68307638168335, "identity": 0.6666420698165894, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b6a8208322ebe5771e7148bb74a448bebb31c38c254721eced59019b84f97075.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "b6c2c1b0a3ec919989a363b0b8a206f831fc6db072b9efabe228f5791778977f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8c5b231e38916157abf509cdb42620ab82b5f816822d8efaeb4d7c3c64616a52", "generation_seconds_batch": 16.101050292956643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4861111640930176, "target_z": 0.6445313579687982, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.5124174952507019, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b6c2c1b0a3ec919989a363b0b8a206f831fc6db072b9efabe228f5791778977f.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "b766643d07abc0a5e2a612f42e2833122f4fed7bd10a16aaf0c640af5a02487a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6fa778eb09e7c8c69dc450971289dad8bbcc5b4aa646af9f630a0f2bb7802f31", "generation_seconds_batch": 24.520703764050268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0760884284973145, "target_z": 0.3222656789843991, "ce": 6.292664527893066, "genuineness": 4.494794845581055, "identity": 0.658515453338623, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b766643d07abc0a5e2a612f42e2833122f4fed7bd10a16aaf0c640af5a02487a.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "b7db8a283abafdfc9c5f77aa6aabb36354530be4ff807343848347829c68ff04", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "561f8d194852e09dc8ce88a82c4ac411001bee6cb4e1e6ffefb941f07290ba4e", "generation_seconds_batch": 16.056258635013364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.355855941772461, "target_z": 1.3281248931291896, "ce": 6.284721851348877, "genuineness": 4.582659721374512, "identity": 0.7042598724365234, "hypothesis": "De deur 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_S_MONO_reg/b7db8a283abafdfc9c5f77aa6aabb36354530be4ff807343848347829c68ff04.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "b7f5e7f961f0936cb71c1a136945e5c880244bf22aa7cb7a52703c88220ed99e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e721d165191842af2de97fb0e123a320631f2cdb4aee23cb585128c86f3e6e79", "generation_seconds_batch": 27.10434698499739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6053903102874756, "target_z": 0.73828122035433, "ce": 6.0305562019348145, "genuineness": 4.65797233581543, "identity": 0.7549748420715332, "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_S_MONO_reg/b7f5e7f961f0936cb71c1a136945e5c880244bf22aa7cb7a52703c88220ed99e.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "b8141833f03e0c12ea7a8453005e5d7362cdd59a2fe916157e4a72bf55e3f932", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "27bee051609aba709471201dd24cb6b5e0133a625812d2f526b91e836d819769", "generation_seconds_batch": 24.023969881003723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.594424843788147, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b8141833f03e0c12ea7a8453005e5d7362cdd59a2fe916157e4a72bf55e3f932.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "b86ff97553d46c14598e9336b17e67adb433930db3582650cfec58aaca30680a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f59e824a507fb2754ff01cb06f04e56e346a29bfbb05df54c45ed5bbc8cd3844", "generation_seconds_batch": 17.052611658000387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 6.252951145172119, "genuineness": 4.494794845581055, "identity": 0.6410081386566162, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b86ff97553d46c14598e9336b17e67adb433930db3582650cfec58aaca30680a.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "b93c4bc49d00576c7782d82fbc1154c81836aaf8aaf7b6968ad111da2282f636", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6709c5f3f1dcd60ba8e330e47fb8d63b504ed0906a604011564d8193b49cf90e", "generation_seconds_batch": 15.550295089022256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8872294425964355, "target_z": 0.1738281314049287, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.6535243988037109, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b93c4bc49d00576c7782d82fbc1154c81836aaf8aaf7b6968ad111da2282f636.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "b99bdf8f3bae52cd3ad333fa4ea61a46a1e66171992d6fb3cf65514617cfa292", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cb46ad67f8296d6eee48c180b2986d0ccf3c8ff8e0148dc5b644b08e7fe9900c", "generation_seconds_batch": 23.30289683502633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.109982967376709, "genuineness": 4.281409740447998, "identity": 0.5956009030342102, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b99bdf8f3bae52cd3ad333fa4ea61a46a1e66171992d6fb3cf65514617cfa292.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "b9f6b138ec18cbbd62236884ea0989641c43514ec7b423e110753fff1e0f10e9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a006bbdec4ee4662c87c064d30be263abd63c3735b0650b9f422e705029605c5", "generation_seconds_batch": 22.735376843018457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 6.197352409362793, "genuineness": 4.337893962860107, "identity": 0.7862691879272461, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b9f6b138ec18cbbd62236884ea0989641c43514ec7b423e110753fff1e0f10e9.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "b9f76816e858b5d6c88f318bd29e2fe6bcaa38fa265fd3bf40f5ee48f2a40c0e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "03aa21e9ba4b1ec91a25487f7c28517fceb086a7250d57a97651ec80204ff171", "generation_seconds_batch": 24.572302518063225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.09409761428833, "genuineness": 4.607763767242432, "identity": 0.8074889183044434, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/b9f76816e858b5d6c88f318bd29e2fe6bcaa38fa265fd3bf40f5ee48f2a40c0e.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "ba09e0715e398cafe0726d07f62459bc995d6f8c13950c28ecc3886bac42f140", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7aff49db08b268629f88f886934d2b3cdd2b6f0b19981f4f630d7523a43a28e4", "generation_seconds_batch": 20.371678151888773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.45844438672065735, "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_S_MONO_reg/ba09e0715e398cafe0726d07f62459bc995d6f8c13950c28ecc3886bac42f140.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "72465f7334a244582dfc09d30ffcfe1a7e35de39d6da2bbaf56e2ea8606fd02a", "generation_seconds_batch": 19.658429977018386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5556907653808594, "target_z": 0.6992188557728227, "ce": 6.078212261199951, "genuineness": 4.432034492492676, "identity": 0.776500940322876, "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_S_MONO_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "ba9a635d2ba33184b02b224bd9fc47ec11cbe2f40b719534ad795a8ae73c2911", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dc58c458e789e4d84c4ce26a53d772d02d2e6af5d07c4afc639c3944db02606e", "generation_seconds_batch": 22.43877272796817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3022220134735107, "target_z": 0.49999995315252144, "ce": 6.1894097328186035, "genuineness": 4.149612903594971, "identity": 0.6095454096794128, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ba9a635d2ba33184b02b224bd9fc47ec11cbe2f40b719534ad795a8ae73c2911.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "bac8b015a8b9b5e0ad7a76f26fbc3dc90f53efe9aff16181e75ef8e052f61bd8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3244be247781121ce2c71b31a8349108cc1388605ebd95142440f522aac56de2", "generation_seconds_batch": 18.73556381301023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 6.141753673553467, "genuineness": 4.6203155517578125, "identity": 0.7151608467102051, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bac8b015a8b9b5e0ad7a76f26fbc3dc90f53efe9aff16181e75ef8e052f61bd8.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "bb0c40721aab9abb6d608f8e366812df1148848bf950e8a0a6b1226fc45d0932", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0e3e698b117014b5a80c319081030b9c89f4d91805aaa967c54a67491dff8bb3", "generation_seconds_batch": 16.237013614038005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6501200199127197, "target_z": 0.7734374421726437, "ce": 6.117925643920898, "genuineness": 4.168440818786621, "identity": 0.5846790075302124, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bb0c40721aab9abb6d608f8e366812df1148848bf950e8a0a6b1226fc45d0932.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "bb0d5649bdff25af38105f689c9f1a5c43499928e87e48f042a0f4adcfc61fce", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45f479d67ab242f5824b1b8720b67df0a4de534c26837d9559aaeb0ced5197fc", "generation_seconds_batch": 27.43196982401423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.998018264770508, "target_z": 1.0468751185826801, "ce": 5.657250881195068, "genuineness": 4.369274139404297, "identity": 0.7767166495323181, "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_S_MONO_reg/bb0d5649bdff25af38105f689c9f1a5c43499928e87e48f042a0f4adcfc61fce.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "bb6d6ca2282168a4bd80b99dd4b2f742143a686d165523863a56a47e36550e22", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "37dc463bcb509198f411723df14cb51bc6d04dc4b297fed792f22cb09ad9aa41", "generation_seconds_batch": 21.72323978599161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.32603645324707, "target_z": 1.3046876617701995, "ce": 5.855817794799805, "genuineness": 4.607763767242432, "identity": 0.49009910225868225, "hypothesis": "Te que he encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.23529411764705882, "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_S_MONO_reg/bb6d6ca2282168a4bd80b99dd4b2f742143a686d165523863a56a47e36550e22.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "bba2a01d4697f1ef247ccdb29010568e147d5df4bff4307e14d88865d71d0c5c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "236ec440320cc6d56f1086ab241cd6b9a878bdfede9d0133cba5f10c3b5236a9", "generation_seconds_batch": 18.939092115033418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.419747352600098, "genuineness": 4.557555198669434, "identity": 0.41487130522727966, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bba2a01d4697f1ef247ccdb29010568e147d5df4bff4307e14d88865d71d0c5c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "bc50da5e84a4a7c35aab26aa34f5f5aeace690c76d55da284bb6cc5ad55ed161", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "42d64d8af2d3748beb519ae58ed1b0fc7081e7e384bad7160087e7c99d581c1e", "generation_seconds_batch": 24.331913403002545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.137177467346191, "target_z": 1.156250114190729, "ce": 6.0146708488464355, "genuineness": 4.570107460021973, "identity": 0.7072243094444275, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bc50da5e84a4a7c35aab26aa34f5f5aeace690c76d55da284bb6cc5ad55ed161.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "bcea0da8a90a28609bb8f525d9a4bb224db5a6d8273178eb3940d617a6a584ed", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfbdf2664b1679be68a0e8929479e5c9f975ab9c38c7d08441f9870379418cea", "generation_seconds_batch": 16.95255221403204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196816921234131, "target_z": 1.2031249516885376, "ce": 6.3641486167907715, "genuineness": 4.6705241203308105, "identity": 0.6970483064651489, "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_S_MONO_reg/bcea0da8a90a28609bb8f525d9a4bb224db5a6d8273178eb3940d617a6a584ed.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "bcf0dffba0194e005a85c4434be5bd30c64a2df8bbeae7a8bd442856d542198b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "04ca4eb302141308e9f3126a378f3e517d8c33c7775b9619fbb4aa3fd224666b", "generation_seconds_batch": 17.703343112021685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7544894218444824, "target_z": 0.8554686888786803, "ce": 5.919358730316162, "genuineness": 4.068024158477783, "identity": 0.4426068067550659, "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_S_MONO_reg/bcf0dffba0194e005a85c4434be5bd30c64a2df8bbeae7a8bd442856d542198b.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "bd17519769e51f3ea6ae0b403a2f39021dbbf89c54117c7e6118d9d6fbd0a1fc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "656f17e3ab7886c4a74e6b44da832c33beebddf0169184f494b404659fb5869c", "generation_seconds_batch": 22.870237127994187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.7460031509399414, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bd17519769e51f3ea6ae0b403a2f39021dbbf89c54117c7e6118d9d6fbd0a1fc.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "bda4d98cff06dab9209c036b8b642bb4a8c47fd45cba13f343e50155de782bbf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "251fbb6d722d340eee27d8eacf4c27ceceaf10951c2cde6ac2d09fb3cfe77504", "generation_seconds_batch": 17.275074098957703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.332377910614014, "genuineness": 4.20609712600708, "identity": 0.6547689437866211, "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_S_MONO_reg/bda4d98cff06dab9209c036b8b642bb4a8c47fd45cba13f343e50155de782bbf.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "bdc1edbade8c7cc97c001d07045ae03d9ce913dad7b9200efbafa1ef683192f4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ceb83b3dacc9cf3abd0d5c4c1222dfc0e56e3fa4bf49caf5901e868a30e31835", "generation_seconds_batch": 17.727117046015337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.938171625137329, "target_z": 0.21386717221967008, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.6751742959022522, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bdc1edbade8c7cc97c001d07045ae03d9ce913dad7b9200efbafa1ef683192f4.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "be1b4370d8f7c88701574e116b7154341b6d8488b1f05e4fdaceecb57ba9565e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f2dbd1d7ea4610974c87465cb1e33874188075bb08a137f104d9ba6e0847c5e5", "generation_seconds_batch": 22.410965457907878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.047718048095703, "target_z": 1.0859376705541017, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.7988365888595581, "hypothesis": "La porte 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_S_MONO_reg/be1b4370d8f7c88701574e116b7154341b6d8488b1f05e4fdaceecb57ba9565e.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "be3453de154f2ff3164c669a3b3665e7737db7bed163f9cda954b8246f8bf88a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f319032117a543357c02233556e80a6a1cc1e2ead8e227254e1869c4d467c466", "generation_seconds_batch": 24.50797356897965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.077537536621094, "target_z": 1.1093749019130918, "ce": 5.6532793045043945, "genuineness": 4.143336772918701, "identity": 0.48623695969581604, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/be3453de154f2ff3164c669a3b3665e7737db7bed163f9cda954b8246f8bf88a.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "be8fa75a5886e8168fac190d9e5e7277bb69c70553b847da462da20f8b4471b9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "643357114f904b978ee9fb352f6ca2fd8308346decc47a18bfe0bed266321c64", "generation_seconds_batch": 23.860926738008857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 5.919358730316162, "genuineness": 4.507347106933594, "identity": 0.8163183927536011, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/be8fa75a5886e8168fac190d9e5e7277bb69c70553b847da462da20f8b4471b9.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "bebf7974431576ecfac33d5e70f17f48fb84af4bba17171588a0c157513242f2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8ff0973abdc19dbb9fb7484d1d585410c9079b07d9ba572a2862848295c9a5eb", "generation_seconds_batch": 23.4067057230277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.1020402908325195, "genuineness": 4.557555198669434, "identity": 0.7541253566741943, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bebf7974431576ecfac33d5e70f17f48fb84af4bba17171588a0c157513242f2.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "bedf378e8c42a3e7ef38d6b24a53f8ba300b25075d008caf80d37c30b65d1584", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "618b696670989a0ff6d7e245861e98bbe1fa236d0c7739474f52b52e9902cbe6", "generation_seconds_batch": 25.902563796029426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.149696350097656, "genuineness": 4.1998209953308105, "identity": 0.5021038055419922, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bedf378e8c42a3e7ef38d6b24a53f8ba300b25075d008caf80d37c30b65d1584.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "bef0e5f4cfab101fc9e2972d2061fd005d83c095dfcf1d4b8e87514d0fc63fad", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9877b3f0622bb5bbec17cc6bf47443aa6090d2d8e7aaadc32f8ecb743661c21e", "generation_seconds_batch": 15.573615973000415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7816176414489746, "target_z": 0.09082039585557222, "ce": 6.165581703186035, "genuineness": 4.287685871124268, "identity": 0.5165870785713196, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bef0e5f4cfab101fc9e2972d2061fd005d83c095dfcf1d4b8e87514d0fc63fad.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "bef14b80af0a48745472b5e76ba68dc8dab8e25b513b2078e8d1e17ae94dc01a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "531180a538d7b61a9c86cb5d699b6c5d01b3555cd109ecd5046b66fa40f650fd", "generation_seconds_batch": 23.20635403401684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6004204750061035, "target_z": 0.7343750775911364, "ce": 6.125868320465088, "genuineness": 4.595211505889893, "identity": 0.734498918056488, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bef14b80af0a48745472b5e76ba68dc8dab8e25b513b2078e8d1e17ae94dc01a.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "bef383a77e8a4d2c873cf076370954e8814b4c30aca5abf8ee80c78d8620d6f5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9cb87e21ea2f23edf141687c2ad0db201ed375489b02274e4e304b7bb0c13e3f", "generation_seconds_batch": 24.065826344070956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9184985160827637, "target_z": 0.98437496047244, "ce": 6.086154937744141, "genuineness": 4.068024158477783, "identity": 0.6545541286468506, "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_S_MONO_reg/bef383a77e8a4d2c873cf076370954e8814b4c30aca5abf8ee80c78d8620d6f5.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "bf3075e34d8cf07e03e4f7cf57f7a1b99188bf1b816129603ee8f9fbbe893229", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7091cb39d509b51c827f57de4e9d937b7b16f85bea2c44b580cd12575de613cb", "generation_seconds_batch": 27.780892102979124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.356205940246582, "genuineness": 4.030367851257324, "identity": 0.2596852481365204, "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_S_MONO_reg/bf3075e34d8cf07e03e4f7cf57f7a1b99188bf1b816129603ee8f9fbbe893229.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "bf7c89295f5c94cc2139d6f4c804fbf74047f040f3e662e1ebadf23c8a7688ae", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "966b0dd9b8d3591a01903295142a566080358818b73fcd2f4fcdd4237645f08b", "generation_seconds_batch": 22.410965457907878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.475135326385498, "target_z": 1.4218749429046356, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.8833814859390259, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/bf7c89295f5c94cc2139d6f4c804fbf74047f040f3e662e1ebadf23c8a7688ae.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "c00667624151b61dae908a0ec8c4c63e7b88616d747d496bdec8e5c20d4a4792", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "48a83ec48a1be2c7b2861baf4e9780977a3ebeffb57f49f22577dae52fff1613", "generation_seconds_batch": 22.161610365030356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4635393619537354, "target_z": -0.15917967441564562, "ce": 6.181467056274414, "genuineness": 4.325342178344727, "identity": 0.5233010053634644, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c00667624151b61dae908a0ec8c4c63e7b88616d747d496bdec8e5c20d4a4792.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "c01cd4aac07a2a51281055cf629c895701a3e9e8db4d1fceadfaac827890e70f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b0ced99462f02ae2f31801b7c6667112178b80a95959d3416c73a2c5d665f1f5", "generation_seconds_batch": 19.76527239906136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7147297859191895, "target_z": 0.8242187972134745, "ce": 6.23706579208374, "genuineness": 4.394378662109375, "identity": 0.6521158814430237, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c01cd4aac07a2a51281055cf629c895701a3e9e8db4d1fceadfaac827890e70f.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c030d933bc6c3e5dc61846c5fa167d204533cf64e60489052470f4ced975ca29", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Someone moved the little blue chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a959ba93b5576023ba566dc3d84c31e9da51c344fc34c8b5ed77ac7319974013", "generation_seconds_batch": 16.958772251033224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.372091293334961, "genuineness": 4.168440818786621, "identity": 0.2459000200033188, "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_S_MONO_reg/c030d933bc6c3e5dc61846c5fa167d204533cf64e60489052470f4ced975ca29.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "c0bb74c9a42d88a770a65e785fef2b1a7eb2fe899e8b888fb04903ea43108ad8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "36c6818d40a3038c955cff8e1efab3b3aee217aae66b779269ae90afc74f8b3a", "generation_seconds_batch": 23.929362587979995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.654054641723633, "target_z": 1.5625002049577188, "ce": 6.1020402908325195, "genuineness": 4.582659721374512, "identity": 0.6717960834503174, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c0bb74c9a42d88a770a65e785fef2b1a7eb2fe899e8b888fb04903ea43108ad8.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "c0c5c7f28a453b7d66292a4dfce318351fe27b5a47166f8b89280ed23b550e7c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b74df23479a9685b8dc6cd3c578f3ed432ff29787a969d0446c7139ff741193", "generation_seconds_batch": 23.721486566006206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.077537536621094, "target_z": 1.1093749019130918, "ce": 6.324435234069824, "genuineness": 4.325342178344727, "identity": 0.35340625047683716, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c0c5c7f28a453b7d66292a4dfce318351fe27b5a47166f8b89280ed23b550e7c.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "c1332243344a6b5242755c22c213abe83546931f597fc120fd35463fabd46058", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4ee077684d1dbbbb4bf7ac523cf25ddd7e99e329c0bc9a4b227680835b14ff84", "generation_seconds_batch": 22.675014531007037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7296395301818848, "target_z": 0.8359374128929695, "ce": 6.141753673553467, "genuineness": 4.268857479095459, "identity": 0.8339222073554993, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c1332243344a6b5242755c22c213abe83546931f597fc120fd35463fabd46058.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "c133c548a47e35df827d0674219c2a43958147b785e90faeeab786806cc0744d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fa629512b34f3450c980e71679233f9616fbe1416abb1fce50ade3e54e3cc39", "generation_seconds_batch": 18.363242020946927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 5.673136234283447, "genuineness": 4.20609712600708, "identity": 0.7187517285346985, "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_S_MONO_reg/c133c548a47e35df827d0674219c2a43958147b785e90faeeab786806cc0744d.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "c201de76644ba8926c3ad8d1ba4c0a9c5c79089dfb2e9e868b0c1060d01fec49", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "da62fa7769e4d2bb1db6c74940de29f7f28415b6fe09d9d1972d7b315f1f0419", "generation_seconds_batch": 22.48715153406374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.351921796798706, "target_z": 0.5390625051239429, "ce": 6.205295085906982, "genuineness": 4.024092197418213, "identity": 0.4333205223083496, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c201de76644ba8926c3ad8d1ba4c0a9c5c79089dfb2e9e868b0c1060d01fec49.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "c25468418813539ae27fb0c0a1f9f417df7304cee1fde9d1bc5f54e0ba9305cb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "fd5fc79340fcb9c094a19d575c45f64b614db4de7625b0e87caebb7b63a4f84b", "generation_seconds_batch": 22.738490099902265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2351274490356445, "target_z": 0.4472656204250509, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.6741106510162354, "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_S_MONO_reg/c25468418813539ae27fb0c0a1f9f417df7304cee1fde9d1bc5f54e0ba9305cb.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "c267716d6169fc185ea547c940254ff2fd6d9104219aaabefc5c50df70b90f85", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96bb7c7cbcd30a9eafdd5ce641852cc5dc73ece5b82236af7429fed4fba6b667", "generation_seconds_batch": 16.780870794085786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.998018264770508, "target_z": 1.0468751185826801, "ce": 6.419747352600098, "genuineness": 4.457139015197754, "identity": 0.41022560000419617, "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_S_MONO_reg/c267716d6169fc185ea547c940254ff2fd6d9104219aaabefc5c50df70b90f85.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "c27cacc9bfc638bfce1c954c1275cfbfcbcf9fa66f6c8057535eade28e87e966", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "490d11f9608513ef78cc2f896c39c2239cb91dedc895441b1d8e1144e70af880", "generation_seconds_batch": 20.657919200952165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.853888988494873, "target_z": 0.9335937928215233, "ce": 6.3641486167907715, "genuineness": 4.413206577301025, "identity": 0.6100437641143799, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c27cacc9bfc638bfce1c954c1275cfbfcbcf9fa66f6c8057535eade28e87e966.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "c2fea22f8bedd57ddda23e1783e00dfefd5ab9e9981a18eb7b9abbd8ff1df40f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9ad69378cb1903b7f9b9a9da2871dff2d17cd04f92e8ffd27539c5fd9da54714", "generation_seconds_batch": 21.682594570913352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 6.395919322967529, "genuineness": 4.118232727050781, "identity": 0.590274453163147, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c2fea22f8bedd57ddda23e1783e00dfefd5ab9e9981a18eb7b9abbd8ff1df40f.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "721ed9c8841b7a8eeec2570171336b0821a44618efb7ae280d03e1889889728b", "generation_seconds_batch": 16.678481868933886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.341981887817383, "target_z": 0.5312500322076416, "ce": 6.117925643920898, "genuineness": 4.168440818786621, "identity": 0.6553177237510681, "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_S_MONO_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "c3e60e0ef137a352929df570980922664a4f9d6d9c250ce50c82657f386e33cf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "da37f87150cbe0713d82464c7e97b54eba194b67a1e99e07981e99bfcee5c212", "generation_seconds_batch": 26.90478867595084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3320419788360596, "target_z": 0.5234375592913401, "ce": 6.260893821716309, "genuineness": 4.68307638168335, "identity": 0.8265234231948853, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c3e60e0ef137a352929df570980922664a4f9d6d9c250ce50c82657f386e33cf.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "c427a3ebe78e8df4148c58300956c882b65133e86d4b93b2f08f23e7620d1115", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3135d01aa6ea7f924239f699d474f2d588bae542652d417e99c07de6f977ffde", "generation_seconds_batch": 16.14919034903869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.2767791748046875, "genuineness": 4.005263805389404, "identity": 0.39816924929618835, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c427a3ebe78e8df4148c58300956c882b65133e86d4b93b2f08f23e7620d1115.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "7a414591414a18cadd373b2f3a7a891fce6ee435b98d95202d86e26852a14772", "generation_seconds_batch": 16.48467895796057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.630965232849121, "target_z": -0.027587917960320747, "ce": 6.078212261199951, "genuineness": 4.507347106933594, "identity": 0.5261448621749878, "hypothesis": "Iemand heeft dit kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "c49cbb42d95a5f5e47e51adbe012bf883fe5d747d75f322bd07f83f5cd38a8b1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6b6fb74e60384167d92b903fcfdf6391ccae0737af23ecd602dfe582fcda19a4", "generation_seconds_batch": 23.23275757499505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.324435234069824, "genuineness": 4.406930446624756, "identity": 0.5651960372924805, "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_S_MONO_reg/c49cbb42d95a5f5e47e51adbe012bf883fe5d747d75f322bd07f83f5cd38a8b1.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "c4a1c76de5fb8c674de2ce4ba0e8492512934972f848f5fca5a4d9e914e7e4dc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "011c8c6684e65bace3ff9f375a3bbe8eff290fd5279149d5d02e4fa30988a4f3", "generation_seconds_batch": 24.753312492044643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.7620395421981812, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c4a1c76de5fb8c674de2ce4ba0e8492512934972f848f5fca5a4d9e914e7e4dc.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "c4a3bab6246f6f08f9eed5809c1420bc0311d7ff734deda1cea634e2b9301a56", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c4afc35a918b8e3ad0406879313729a1bc4d4b9122d51f3bb7df2a1fc9651185", "generation_seconds_batch": 22.161610365030356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.893441915512085, "target_z": 0.1787109504013564, "ce": 6.300607204437256, "genuineness": 4.68307638168335, "identity": 0.5953845381736755, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c4a3bab6246f6f08f9eed5809c1420bc0311d7ff734deda1cea634e2b9301a56.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "c4bd10bf0fa1b8bd3587c5159531c822d93ea226a795b85394a5fc4b35cf8dd2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7604acd0ca1959c7545eac44ed7a6610cbdce2089745eec7895617624bc96124", "generation_seconds_batch": 16.952595390030183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7107954025268555, "target_z": 0.035156221818313675, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.3581814169883728, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c4bd10bf0fa1b8bd3587c5159531c822d93ea226a795b85394a5fc4b35cf8dd2.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "c4f275eca48499b072fee64c6fca144bae7900a9341e96b60b9546b6db8c55a3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f13043340ee8b7215cd8ec92cd573186db8a8d5c7f0b36d75c441f49058e8740", "generation_seconds_batch": 22.40682736504823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.221180438995361, "genuineness": 4.821148872375488, "identity": 0.7634651064872742, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c4f275eca48499b072fee64c6fca144bae7900a9341e96b60b9546b6db8c55a3.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "c4fd91a58a7cfb37cbb958d0ba166c565cb58ae2903749a352fbef1e3ced4f72", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9fae0cf01d1c037383680d12dd6545e03073c322fb67f37cb6ec94acb157faf9", "generation_seconds_batch": 17.39460976794362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5399527549743652, "target_z": -0.09912101949857645, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.49184373021125793, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c4fd91a58a7cfb37cbb958d0ba166c565cb58ae2903749a352fbef1e3ced4f72.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "c56f6347a39424a38a3a80ec0d41861f0441c585b5cbfdb84263e54f448d03e8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9c49703aa1228119d0d291293ae57452c02e95f9acdcf701af6438f32d22ece9", "generation_seconds_batch": 25.03464772098232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7246696949005127, "target_z": 0.832031270129776, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.27504390478134155, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c56f6347a39424a38a3a80ec0d41861f0441c585b5cbfdb84263e54f448d03e8.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "c5a5c2e884cda26d78e637b0fca366b55330ad1c4822375a4b7a825324431d18", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1a5507c2dab548d0ceed3aee345c307db9a84ed3c828861f8b2868346fc72e39", "generation_seconds_batch": 23.13551482197363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.93837833404541, "target_z": 0.9999999063050429, "ce": 6.300607204437256, "genuineness": 4.118232727050781, "identity": 0.6086027026176453, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c5a5c2e884cda26d78e637b0fca366b55330ad1c4822375a4b7a825324431d18.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "c5d68dab9b3a72bcd3b66639a19b837db6ca7cbcc066e24390619efbb35e4e33", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d68066ff5f146adbc55fa1ab962c7c7533b4a8190a0e5553c208454ab501723f", "generation_seconds_batch": 27.431504456093535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 6.09409761428833, "genuineness": 4.65797233581543, "identity": 0.5568600296974182, "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_S_MONO_reg/c5d68dab9b3a72bcd3b66639a19b837db6ca7cbcc066e24390619efbb35e4e33.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "c6124140f920aa357829586a28e20d6db0744b956517a57d2311ded1548d1454", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0343e02e4fc43072cb70f192eb2a1ee52f0e6f3bb9556219f7a99bfc303cb41a", "generation_seconds_batch": 17.257368410006166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.268836498260498, "genuineness": 4.293961524963379, "identity": 0.30029013752937317, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c6124140f920aa357829586a28e20d6db0744b956517a57d2311ded1548d1454.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "c651565c608bdb4fab930294e16ce28548ceb929f32f18ebda2a374abbc22996", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b360a72a1227293214abf1949daa69a94262e239ed9c8c61aec15ae56953d72b", "generation_seconds_batch": 17.253830702975392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8748044967651367, "target_z": 0.16406249341207332, "ce": 6.252951145172119, "genuineness": 4.268857479095459, "identity": 0.6629767417907715, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c651565c608bdb4fab930294e16ce28548ceb929f32f18ebda2a374abbc22996.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "c67cb57803ef82ff73cd858dc799d20e75978495b0ec3c2dc65910e48f16a749", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0b72c1a0efa2303b17cd54252925058b8e1c5ce4842a8b36c7b1ab520521ab74", "generation_seconds_batch": 24.186107353074476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.131171703338623, "target_z": 1.937500029279674, "ce": 4.928510665893555, "genuineness": 4.093128204345703, "identity": 0.5228902101516724, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "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_S_MONO_reg/c67cb57803ef82ff73cd858dc799d20e75978495b0ec3c2dc65910e48f16a749.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "c6f8dbfd4f5078d15b3a0850ccfe9385683d60db27667fc89395b0bc92367fad", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "010dd0276e6dc69a38e5eb5bf650d839b5ff3a25748b1841d29c8648fe697ef9", "generation_seconds_batch": 27.919004468014464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.550720691680908, "target_z": 0.6953125256197148, "ce": 6.1020402908325195, "genuineness": 4.024092197418213, "identity": 0.6634678840637207, "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_S_MONO_reg/c6f8dbfd4f5078d15b3a0850ccfe9385683d60db27667fc89395b0bc92367fad.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "c70c8a4e9a54c66ccd360364fd2c74e8ee506a5c661875019e129a837979233b", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40a0137280c1b83c5ff0026791fd0f7dcdfd619a7cabd14a28322528db980347", "generation_seconds_batch": 16.744006821070798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 6.4515180587768555, "genuineness": 4.011539936065674, "identity": 0.2457572966814041, "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_S_MONO_reg/c70c8a4e9a54c66ccd360364fd2c74e8ee506a5c661875019e129a837979233b.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "9bf83c80eedfb3e7aa9919b3bcfcf3d050ab348ebc0cd23fdebb3df520aed9cf", "generation_seconds_batch": 20.380800051032566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6635806560516357, "target_z": -0.001953165076553933, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.5940920114517212, "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_S_MONO_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "c8509c277fcd2c72ec890e13ed7fa896fd432758018439c271b9009198440348", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "84301aab6ad506951f0337ac47bca5dbea3c44fda6a8d1d0d635848630c3f0a4", "generation_seconds_batch": 16.285176000907086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602698564529419, "target_z": -0.04980467880759675, "ce": 6.173524379730225, "genuineness": 4.519899368286133, "identity": 0.8426627516746521, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c8509c277fcd2c72ec890e13ed7fa896fd432758018439c271b9009198440348.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "d7b29fff2aa68a8ad9df502b38dac064f18064005db4fe2c5ba1b9cd1ca0fb52", "generation_seconds_batch": 14.967284665093757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.720114231109619, "target_z": 0.04248054400791235, "ce": 6.372091293334961, "genuineness": 4.268857479095459, "identity": 0.2078145295381546, "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_S_MONO_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "c8988176cf3c421ed3affd0486ae2911cfca92b532ca46c8da5a990450b3ea40", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b6a8a8c4dab34b6c59be93f83491222f574ba5e20c42e113b18448d49ae66f5f", "generation_seconds_batch": 17.699867147020996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 5.974957466125488, "genuineness": 3.973883628845215, "identity": 0.5367475748062134, "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_S_MONO_reg/c8988176cf3c421ed3affd0486ae2911cfca92b532ca46c8da5a990450b3ea40.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "c8af1219a51bce2e0b27977e5c47fa9def6d50b1a586ed51ef43dc5fac1fd446", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "851a6ded881509b5a011eea5afb58c04c8615cb4fdef3688668bb220ac1fc0bf", "generation_seconds_batch": 23.67651411902625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.221180438995361, "genuineness": 4.758388519287109, "identity": 0.6718896627426147, "hypothesis": "He encontrado, la cajita estaba junto a la ventana. Vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c8af1219a51bce2e0b27977e5c47fa9def6d50b1a586ed51ef43dc5fac1fd446.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "c903a7b72aca8a6f61063c7a7a2978015f3542c593fbdc6ec7906a824d72481d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e70f0313afb2fb465f8e6090ebb6edffdc8ee7e86be6ce77c46c98c124221bc4", "generation_seconds_batch": 18.32607819302939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2301573753356934, "target_z": 0.44335929027194304, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.5428671836853027, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c903a7b72aca8a6f61063c7a7a2978015f3542c593fbdc6ec7906a824d72481d.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "c9759ecb02c1d14f857bbfd246b40a82f612f331e1871ec313f3271d594eb0f5", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0a3f22dd962eb436de09a7dde69f6a2e06ef1fd89ac41b67dce5a2bcefcfb441", "generation_seconds_batch": 17.015110059990548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994083881378174, "target_z": 0.2578125431875193, "ce": 6.340320587158203, "genuineness": 4.356721878051758, "identity": 0.39748769998550415, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/c9759ecb02c1d14f857bbfd246b40a82f612f331e1871ec313f3271d594eb0f5.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "ca1f5218ec6bfef38eb5deaed2d5d98df81f20b4787369cf0fdf06ed9859600f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "20cb95977015f58bf879d8288f02b4474e1e4a5342d34da226a46d781fe24179", "generation_seconds_batch": 23.073613230953924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.127237319946289, "target_z": 1.1484374538845132, "ce": 6.38003396987915, "genuineness": 4.306513786315918, "identity": 0.6168364882469177, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ca1f5218ec6bfef38eb5deaed2d5d98df81f20b4787369cf0fdf06ed9859600f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "cac3db25fc1856bba3f98558f8bb4f80a1bfc15479645aa7f848f13fbff1343d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7b043cdd02f1583e2e9f04f8369c0782b8b7363e3ead95a0ef2247065130d1ce", "generation_seconds_batch": 24.105277594993822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7843093872070312, "target_z": 0.878906295017499, "ce": 6.316492557525635, "genuineness": 4.319066047668457, "identity": 0.6231606006622314, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cac3db25fc1856bba3f98558f8bb4f80a1bfc15479645aa7f848f13fbff1343d.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "cb31e5ac17f75fc443288694c298baabc8cfa0bd8fd178cccc97240a51c6d064", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "29c735ad0d66d43b011f40b9f4d2331642ceb74fd6c984d44f89f36d19b34f33", "generation_seconds_batch": 18.134099129936658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2400975227355957, "target_z": 0.4511719505781588, "ce": 6.435632705688477, "genuineness": 4.482243061065674, "identity": 0.39798641204833984, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cb31e5ac17f75fc443288694c298baabc8cfa0bd8fd178cccc97240a51c6d064.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "cb511fa4dec75afa9b40529914e256d1aeafe576ac0731aa59d55f955a66239f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8a230e275ba4f2f72942c689eadab5af989cecdd752eb926794fe3bc2f33c701", "generation_seconds_batch": 22.07985139603261, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4115614891052246, "target_z": 0.5859375300116659, "ce": 6.173524379730225, "genuineness": 4.118232727050781, "identity": 0.4440523386001587, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cb511fa4dec75afa9b40529914e256d1aeafe576ac0731aa59d55f955a66239f.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "cc3463d776cf5152a832ff602229875ca8b28b167a38c725b31a403b0df4391d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "635116a1e2ace73e6fb97312390bec611173215dd5954e510b400e1e65b92153", "generation_seconds_batch": 22.812258829013444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6302402019500732, "target_z": 0.7578124963400408, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.5526418685913086, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cc3463d776cf5152a832ff602229875ca8b28b167a38c725b31a403b0df4391d.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "cca13420f8759f7acf77af97adbd0e7850d9040dc8cb996c56612a1e4f949110", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Escribí el número en el sobre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3906899a98214b5b56e3017ef2b260f6d168a9e2eeac4063134aec89a7fc615", "generation_seconds_batch": 27.403635088005103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.0305562019348145, "genuineness": 4.281409740447998, "identity": 0.7398821115493774, "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_S_MONO_reg/cca13420f8759f7acf77af97adbd0e7850d9040dc8cb996c56612a1e4f949110.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "cd168ae6bbcfe1a4ca8e8912ba765deff473008423b18454c0cba8a43d072942", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5c3d313694c8bd6a204f5908e1eb6f8f242386998da4ba1b1bbc3bfd651a95e0", "generation_seconds_batch": 24.520888195023872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2475523948669434, "target_z": 0.4570312584179063, "ce": 6.006728172302246, "genuineness": 4.268857479095459, "identity": 0.5565887093544006, "hypothesis": "De door staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cd168ae6bbcfe1a4ca8e8912ba765deff473008423b18454c0cba8a43d072942.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "cd7f4089837d59fb1a313906c51a66495362b91fd89bd1c8e5c3208dafd42c86", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "96d0796e8fa8b064ca695f9992166438a2a5413b9200f1f625778850031be5dd", "generation_seconds_batch": 24.527191777015105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.252951145172119, "genuineness": 4.6705241203308105, "identity": 0.7573188543319702, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cd7f4089837d59fb1a313906c51a66495362b91fd89bd1c8e5c3208dafd42c86.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "ce04ab28af881798a80a3e09034d3fac2d5b2fe3a043ee6805a7e8ad516ee7cb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f07ec0ca1e131bba0feb3a8f5b6be52dbc4aa661ff513e21894682c94cba3893", "generation_seconds_batch": 16.96470561006572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4811410903930664, "target_z": 0.6406250278156904, "ce": 5.895530700683594, "genuineness": 4.469690799713135, "identity": 0.7225390076637268, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ce04ab28af881798a80a3e09034d3fac2d5b2fe3a043ee6805a7e8ad516ee7cb.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "ce55b1c21610ddabae69309a7637ead177d3aac682a1730315f5775c44283d99", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b16e4876703401e1976ebb2f15773672fb96ce1d578eab068fe803b84b90c97a", "generation_seconds_batch": 22.739073859062046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.0146708488464355, "genuineness": 3.936227321624756, "identity": 0.650065541267395, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ce55b1c21610ddabae69309a7637ead177d3aac682a1730315f5775c44283d99.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "ce9135e67097e5987981ea4b0a0a4c720eb0e1c260d3823a27bb065a0822090b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "24b1d8aa900b72853d477bf547ec507cff6237398309eebcf49a48c662343902", "generation_seconds_batch": 24.50797356897965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.133810997009277, "genuineness": 3.886019229888916, "identity": 0.6054402589797974, "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_S_MONO_reg/ce9135e67097e5987981ea4b0a0a4c720eb0e1c260d3823a27bb065a0822090b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "ceab84ef3560fc8019bcb7961f68f5ecf2258d4d4774a79b8e8366f8065d0b0c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8467b238c34bad24353628368121ee64d99f884c38a9f2b1b285ea58777bdfc2", "generation_seconds_batch": 17.185016855946742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431829929351807, "target_z": 0.37500001171186964, "ce": 6.284721851348877, "genuineness": 4.419482707977295, "identity": 0.6282650232315063, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ceab84ef3560fc8019bcb7961f68f5ecf2258d4d4774a79b8e8366f8065d0b0c.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "cecf9bc1d98e4e6bb8ab84fac83bccd3fcdd730e78ed470d73d1cc8ac2516f87", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "34b504902c441af5d6380dd9231ab7181a4db76c46b84c9ae73688e39bd89c9d", "generation_seconds_batch": 25.345219173002988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9394140243530273, "target_z": 0.2148436610629899, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.6343439817428589, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cecf9bc1d98e4e6bb8ab84fac83bccd3fcdd730e78ed470d73d1cc8ac2516f87.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "cee8ad3e113b1bf08d7ad601b5e1a6808e295150d55ea3006123173977877477", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ef03b348a4a763a7b4239c1938ad9542412bc0b2b8ac99b5ef5b958935b0d13f", "generation_seconds_batch": 17.685997131979093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0686333179473877, "target_z": 0.31640618375473734, "ce": 6.292664527893066, "genuineness": 3.9801597595214844, "identity": 0.23319484293460846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cee8ad3e113b1bf08d7ad601b5e1a6808e295150d55ea3006123173977877477.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "ceeb69155a44e3bba17752fe860eaf767c4bf7857f383e60b321c206750200b1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f9c10c023278ba6999adaf069da3d36c255c29c08ed5d5553168aefbb4437cab", "generation_seconds_batch": 23.4067057230277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.133810997009277, "genuineness": 4.482243061065674, "identity": 0.6744450330734253, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ceeb69155a44e3bba17752fe860eaf767c4bf7857f383e60b321c206750200b1.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "cf09ad734d198bfaa75c90af150903790d960ef946ed1b19b93d3040d4e98368", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "02f1f77c51b0d0c4f382bbe377312c250b0faa7cbbf35cc4c17afafdd8cb749d", "generation_seconds_batch": 16.968194245011546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8389790058135986, "target_z": 0.921874989752114, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.7044123411178589, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cf09ad734d198bfaa75c90af150903790d960ef946ed1b19b93d3040d4e98368.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "cf29d2554c3b6c94d37eb239b7b31cd2a0cb08ed2c962bb27363e069d71d7e0a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8e457cc6afe706a8e73b341ff3c458cd5b9409a87b2de042f8296e0c3c5a1f3c", "generation_seconds_batch": 23.721486566006206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.157639026641846, "genuineness": 3.9550557136535645, "identity": 0.3648606538772583, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cf29d2554c3b6c94d37eb239b7b31cd2a0cb08ed2c962bb27363e069d71d7e0a.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "70ef74c084c3e982e1a45f10fb88e98846b095e65ae364dd486e87ea45708e7b", "generation_seconds_batch": 16.28890575398691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4910809993743896, "target_z": 0.6484375007319918, "ce": 6.3641486167907715, "genuineness": 4.68307638168335, "identity": 0.5732549428939819, "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_S_MONO_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "cf886404ce26c89b3fd69ed4935ae7baba0ebcb8b57c93b23a0b2ffa89f9f6f9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f10aad761de36cb7aac46ca2f6ec4195fb92096a4bd582095e2b7476bc149ced", "generation_seconds_batch": 26.424286316963844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.348263263702393, "genuineness": 4.419482707977295, "identity": 0.4564075469970703, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/cf886404ce26c89b3fd69ed4935ae7baba0ebcb8b57c93b23a0b2ffa89f9f6f9.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "d049f4473b0d21e015f97ec5e4fc4236b868f3d937b3417f495ff2538be49285", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6663c18b6745d9ef4675a511314139a30dacf6b78b33f42548ca0f15486a281f", "generation_seconds_batch": 23.009457394015044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.381741523742676, "target_z": 0.5624999238728473, "ce": 6.205295085906982, "genuineness": 3.8922953605651855, "identity": 0.17991384863853455, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d049f4473b0d21e015f97ec5e4fc4236b868f3d937b3417f495ff2538be49285.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d08ce47f05da445b2819e70d344a8dbb0a7e7808f414eedac2d7825b8fe3f336", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf702e072996ddf2292d76af22cc75134336c4c620bbe0548c5263375d1a1e70", "generation_seconds_batch": 23.697775250067934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.127237319946289, "target_z": 1.1484374538845132, "ce": 6.4515180587768555, "genuineness": 4.494794845581055, "identity": 0.5318005681037903, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d08ce47f05da445b2819e70d344a8dbb0a7e7808f414eedac2d7825b8fe3f336.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d0df8caebc8ce75f2f6f5b79266e49df9b0399f438e4a93a8c8b92da1397fd9e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c8ce80c479822ecdb51d6ebc1bede401c255870f20639db1f190d6e1aec3f4e2", "generation_seconds_batch": 26.411029117065482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 5.974957466125488, "genuineness": 3.9927117824554443, "identity": 0.7936766147613525, "hypothesis": "La porte 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_S_MONO_reg/d0df8caebc8ce75f2f6f5b79266e49df9b0399f438e4a93a8c8b92da1397fd9e.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "d0e07ff1d42773f6f8273914cc7208968e1998719449fbab2d3518e75bea8e49", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "efd6c81e2597509064c98614891c1dbfff558d748cc84e1b72771f850caad114", "generation_seconds_batch": 23.681950527010486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.197352409362793, "genuineness": 4.432034492492676, "identity": 0.46801361441612244, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d0e07ff1d42773f6f8273914cc7208968e1998719449fbab2d3518e75bea8e49.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "43e1c7a8e201e344c40d550e0584e5ea2594d730120159afaf3271f3320660b7", "generation_seconds_batch": 15.105511959991418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.123303174972534, "target_z": 0.35937506587926676, "ce": 6.292664527893066, "genuineness": 3.7918784618377686, "identity": 0.4785647392272949, "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_S_MONO_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d1afc99d3ed7afcc967357ee1f8913b722171d21bd1f7886c70efed678e5c21a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "13b70674bb416bb6a42b978eb1a90047f1b0a74f1f2bd14a160925cd562a786e", "generation_seconds_batch": 16.250061376951635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1481528282165527, "target_z": 0.3789061544750632, "ce": 6.197352409362793, "genuineness": 4.400654315948486, "identity": 0.636061429977417, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d1afc99d3ed7afcc967357ee1f8913b722171d21bd1f7886c70efed678e5c21a.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "d1d8cdd82f71a2b2148a5db353f564c1a1dda876a4076a98996a525111a97dcf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "38c88e4a25804cb9b9220bbff1de2a73d27db940c309c6a74a33ef46f8c78042", "generation_seconds_batch": 25.46872396604158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9880781173706055, "target_z": 1.0390624582764645, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.6095892190933228, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d1d8cdd82f71a2b2148a5db353f564c1a1dda876a4076a98996a525111a97dcf.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d1f36587018765fe5c860f601293cf7dbe97522856be6aacb894d07c1eb52d9d", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "83c903c593cade59e97356d2fa5edd6a8822997364e479d8105f16afaab753a8", "generation_seconds_batch": 17.183784766006283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.299736976623535, "target_z": 0.4980467880759675, "ce": 6.173524379730225, "genuineness": 4.356721878051758, "identity": 0.7441446185112, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d1f36587018765fe5c860f601293cf7dbe97522856be6aacb894d07c1eb52d9d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "d2104733f3baafcee56b5effd9e75e2118a2ec05e54cfa4a938cc7a1991a7819", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "45a4231c70dd86de829db066ff891e90be11b2c7348220317cd057c9e184a07b", "generation_seconds_batch": 18.58606014493853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2649474143981934, "target_z": 0.47070322656386954, "ce": 6.0146708488464355, "genuineness": 4.243753433227539, "identity": 0.5629254579544067, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d2104733f3baafcee56b5effd9e75e2118a2ec05e54cfa4a938cc7a1991a7819.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "d2118f92a23fbc77eb63317b080dbe3bdb8787b132d68e0308c0884e684fc2e5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "beac96e9b94bd75c672400b766d9b85c665b6e4184e150483520189a7646c4af", "generation_seconds_batch": 22.675085710012354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.6231074333190918, "hypothesis": "De deur 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_S_MONO_reg/d2118f92a23fbc77eb63317b080dbe3bdb8787b132d68e0308c0884e684fc2e5.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "d2200d3eb4a6cbb06d2807e16fb5af8fd85f356ebb499685e68483b0a842ef8b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "704f0e0ae6b626f06007a47211a74fa87ea888db4e7441cdb88fe59b33337be6", "generation_seconds_batch": 22.768882901058532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.7223818898200989, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d2200d3eb4a6cbb06d2807e16fb5af8fd85f356ebb499685e68483b0a842ef8b.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "88519a044614330b19b52283e3c8967da6f43ed529fd58e594ccbb74376e39a6", "generation_seconds_batch": 16.65727922297083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.42647123336792, "target_z": 0.597656145691161, "ce": 6.149696350097656, "genuineness": 4.262581825256348, "identity": 0.6953564882278442, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d24a5f6042b71b35ea72da6d02d14239a300c8d33010d8a04912d993d23c42a1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "39b6f6b59a714a5d1329249e81d6bad0f203d530106bddaf31394f8d83ffad90", "generation_seconds_batch": 16.90731485199649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.723633766174316, "target_z": 1.6171873279819147, "ce": 6.070269584655762, "genuineness": 4.7081804275512695, "identity": 0.8020679354667664, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d24a5f6042b71b35ea72da6d02d14239a300c8d33010d8a04912d993d23c42a1.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "d255018b4d162d8e0a122ef2a4da0c4137243ffc2b9b5ca4e1db2e022cfa73eb", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "50d3d0d54cebc5dfc03052b2dd6f4192b7881c6171e3241492c9d72ec4479efe", "generation_seconds_batch": 22.62317726504989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.22021746635437, "target_z": 0.4355468173556416, "ce": 6.038498878479004, "genuineness": 4.362998008728027, "identity": 0.5362300872802734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d255018b4d162d8e0a122ef2a4da0c4137243ffc2b9b5ca4e1db2e022cfa73eb.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d26b21e4e2dd47fe7772daab1c7f01f9f5fba74e0f09e6c37c06ca5c570f8336", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "afae4be6a80354abeadcefddd35678f1a6ed2768bced6998fb75f34f754f5f1c", "generation_seconds_batch": 22.96240999503061, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898618698120117, "target_z": 0.968750014639837, "ce": 6.38003396987915, "genuineness": 4.425758361816406, "identity": 0.29482346773147583, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d26b21e4e2dd47fe7772daab1c7f01f9f5fba74e0f09e6c37c06ca5c570f8336.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "d29834728cf060a1688fe642ee122a980677c3c933e5773b3191be09e287db61", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cf83489f8e7b8479177f7693cde223f8d4602803ec7488b98a450931ff83cdf", "generation_seconds_batch": 17.0018320819363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.097417831420898, "target_z": 1.1250002225255231, "ce": 5.911416053771973, "genuineness": 4.482243061065674, "identity": 0.7534481287002563, "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_S_MONO_reg/d29834728cf060a1688fe642ee122a980677c3c933e5773b3191be09e287db61.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "d29f0075b80277007bc8c6db2a921e4ee635040c4ccf12f74e08c6d4022cf462", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "5b45e04386a32700504e1e555cfa6395bbb61c7d91cc5817ea3d876e27bb703f", "generation_seconds_batch": 25.365305499988608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.62030029296875, "target_z": 0.7500000234237393, "ce": 6.1020402908325195, "genuineness": 4.118232727050781, "identity": 0.5421262979507446, "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_S_MONO_reg/d29f0075b80277007bc8c6db2a921e4ee635040c4ccf12f74e08c6d4022cf462.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "d2e74b2444c9045edde4491c67dda974c982bcc09b157713d3c48ab3679d703e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4284f2e688cd84ff292c1d9bb9dda386c805cbc20ec1b7add57084f398b7ac8c", "generation_seconds_batch": 27.970281027024612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.415495872497559, "target_z": 1.3750001054068268, "ce": 6.419747352600098, "genuineness": 4.124508857727051, "identity": 0.17610952258110046, "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_S_MONO_reg/d2e74b2444c9045edde4491c67dda974c982bcc09b157713d3c48ab3679d703e.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "d38300bfd8b1711b5795ea3ae27c65c08acbe179b6d561d92d6eb8b80fba15a9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1d1501875ee29ea444cb34ec2762a04e850f586e3f8e55bd2662ef37bffe2156", "generation_seconds_batch": 17.333124459022656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 5.95112943649292, "genuineness": 4.607763767242432, "identity": 0.6512503623962402, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d38300bfd8b1711b5795ea3ae27c65c08acbe179b6d561d92d6eb8b80fba15a9.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "d3c94ab47bb994e4ebdc78d0de31946877a5108e4654430f3e1d1d2779ace99e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e906827d62b427032bb89faaaa8950454d4a89853313ea365bda098771f5faa6", "generation_seconds_batch": 26.017455643042922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.026388645172119, "target_z": 0.2832031270129776, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.5435633063316345, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d3c94ab47bb994e4ebdc78d0de31946877a5108e4654430f3e1d1d2779ace99e.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "d3d878eca487f1ea63feff3744859597e3f4df992d46228cd89ffd471be66ff8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6568d1c49f48ff3e6340f606e8c6c710883cc9064b18e2f7ac249623798aa5ec", "generation_seconds_batch": 22.532765422016382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3370118141174316, "target_z": 0.5273437020545336, "ce": 6.268836498260498, "genuineness": 4.13078498840332, "identity": 0.644305944442749, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d3d878eca487f1ea63feff3744859597e3f4df992d46228cd89ffd471be66ff8.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "2ff274cfc357c671343553574ff2050788e4d84a45b5fc387f1ab685d686f2cd", "generation_seconds_batch": 15.887489561922848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.7105973958969116, "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_S_MONO_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "d40f85f95d8c6f7d5611394efd805b0a5548de6d1cb66de9b08e6e4f3deeea51", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "58a56c66f54157a4f59a532630a633ff042bf08dc77e06272e8868f643969b97", "generation_seconds_batch": 20.549166548997164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.431441307067871, "target_z": 0.6015624758442688, "ce": 6.292664527893066, "genuineness": 4.632867813110352, "identity": 0.6679661870002747, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d40f85f95d8c6f7d5611394efd805b0a5548de6d1cb66de9b08e6e4f3deeea51.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "d46c8e4c7a874670ac1d922fd94c60d57dccd8259f25da5e34ed2c996732f36f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "940fdcec8073bd463c37f32edaace0a2d4fdfddf094c99e5d765be9101a74ca9", "generation_seconds_batch": 23.938277384033427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.375736236572266, "target_z": 1.343750213741621, "ce": 6.292664527893066, "genuineness": 4.406930446624756, "identity": 0.49759000539779663, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d46c8e4c7a874670ac1d922fd94c60d57dccd8259f25da5e34ed2c996732f36f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "d4fc5d556978bf1cfb70fbbbcc6de1a6e16db8af180c1f654769621e052e5788", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7fbd8f58c1ed3247bfcc4ea673a88829ef49b2728abd1ae7b1316a6c6d901d0d", "generation_seconds_batch": 20.780079418909736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6749699115753174, "target_z": 0.7929687181583545, "ce": 6.300607204437256, "genuineness": 4.099404335021973, "identity": 0.6672981381416321, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d4fc5d556978bf1cfb70fbbbcc6de1a6e16db8af180c1f654769621e052e5788.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "d5741565997ea9c8451e1ad01bd170fd7bc307b59d2b08f9e10c879ae5c14045", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4ab8d8a66a0e1a51535f4280b249d7b5216b59404ccef31bd41cd8fdc4ca12d9", "generation_seconds_batch": 26.476698574027978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.853888988494873, "target_z": 0.9335937928215233, "ce": 6.284721851348877, "genuineness": 3.8420867919921875, "identity": 0.44870075583457947, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d5741565997ea9c8451e1ad01bd170fd7bc307b59d2b08f9e10c879ae5c14045.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "d586f947d0ce25b5bee47ffd82825499bdc1819572927b4ee426e6dad072303f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9afe45756ffef43da851f718f80792a70955c277621be8545d6b4939d2176197", "generation_seconds_batch": 23.866501386044547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.7070616483688354, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d586f947d0ce25b5bee47ffd82825499bdc1819572927b4ee426e6dad072303f.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d634f05f53c1714633719ccbf00a236b3037a8c0153eb68c57bda49d3dfb71a9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b92a7eaface80cbc63a0fbdec2f11034c04c6952af12c25d0729d4a69dab031e", "generation_seconds_batch": 28.649957439978607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7196998596191406, "target_z": 0.8281251273665824, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.7614072561264038, "hypothesis": "La porte 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_S_MONO_reg/d634f05f53c1714633719ccbf00a236b3037a8c0153eb68c57bda49d3dfb71a9.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "d67af740de079898b4fc81eb5800bbf86b086a876340c9cf24611476fd2f8c70", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "de9766468657c27070adc72c0008c6b1757804503be77bb3b7af1bcbcc04c8f5", "generation_seconds_batch": 19.996973425964825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7544894218444824, "target_z": 0.8554686888786803, "ce": 6.260893821716309, "genuineness": 3.961331605911255, "identity": 0.40440165996551514, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d67af740de079898b4fc81eb5800bbf86b086a876340c9cf24611476fd2f8c70.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "d67cae00ddab5e25f5ae7feb7f6c23734883fe58dc7f2b60aad4dc6d3d0714b3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dc674bbd3274030293fee5c4fe08721884cc806eb79aa5d29a80a15dbc969907", "generation_seconds_batch": 23.165404631989077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.93837833404541, "target_z": 0.9999999063050429, "ce": 5.95112943649292, "genuineness": 3.9048471450805664, "identity": 0.36177000403404236, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d67cae00ddab5e25f5ae7feb7f6c23734883fe58dc7f2b60aad4dc6d3d0714b3.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d7aa9b468fba2ec31677f66383db9a343b3f77c9ae4efd25f76aa82709f2a009", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6b1e32c8b9739427499426754ab8c33b6e07b8d9eab0ae295ef649ab50cac225", "generation_seconds_batch": 18.102993670967408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.38003396987915, "genuineness": 4.469690799713135, "identity": 0.5577699542045593, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d7aa9b468fba2ec31677f66383db9a343b3f77c9ae4efd25f76aa82709f2a009.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d7ea209e9d0b376b66dd034fd696aca92f347db99cfb2d4bfe3d8c92d9cfa23f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b6eb7f6c4c063f826f08c4c5fc28f332b957848cac576f6b59d138e2ff35cc86", "generation_seconds_batch": 23.988912081927992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.006728172302246, "genuineness": 4.2374773025512695, "identity": 0.6964946985244751, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d7ea209e9d0b376b66dd034fd696aca92f347db99cfb2d4bfe3d8c92d9cfa23f.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d8462e22654cb99fa2c22c83d86beef24d331a80b6353524da0266cabef39060", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "39d1f38345c9541fd7a42a335ca65de04b3f44ae3e4fc5b8c3b9bbfdded91d36", "generation_seconds_batch": 18.41823343699798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.41180419921875, "genuineness": 4.231201171875, "identity": 0.3656759262084961, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d8462e22654cb99fa2c22c83d86beef24d331a80b6353524da0266cabef39060.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "d87777c04a9d98983463a8e7b09b3129b5bd4e1cf32d4f5ee1e1c2e713dfff24", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39efc9c4da787537c251a5dfa9d83c6bfc973c0c660f2292762e7c89a3f1ca4b", "generation_seconds_batch": 27.634806237067096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.465195655822754, "target_z": 1.4140626573782482, "ce": 5.927301406860352, "genuineness": 4.582659721374512, "identity": 0.788668155670166, "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_S_MONO_reg/d87777c04a9d98983463a8e7b09b3129b5bd4e1cf32d4f5ee1e1c2e713dfff24.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d88e012c4d7ff7e1919ce549688fcadd37510565829f44395010a23a93b997d3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "003f96aecef4a5a7483e98a0688bbef4933f90b79c83e1d1f2afc314f8e7cea3", "generation_seconds_batch": 24.602987863007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7097597122192383, "target_z": 0.8203124670603666, "ce": 6.316492557525635, "genuineness": 4.557555198669434, "identity": 0.500118613243103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d88e012c4d7ff7e1919ce549688fcadd37510565829f44395010a23a93b997d3.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d8e3fb8e1303ac959d8974e531698f4969e2f3412020cb5844713460f029c5a4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7853d7a5e0b9a259292a98bcf5597023c7158f7b4dc09501c80716e88165359d", "generation_seconds_batch": 26.424286316963844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.276336669921875, "target_z": 1.2656251097987778, "ce": 6.125868320465088, "genuineness": 4.20609712600708, "identity": 0.4900866150856018, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d8e3fb8e1303ac959d8974e531698f4969e2f3412020cb5844713460f029c5a4.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "d8f60f39c6c3b88a2b5e622ce4ff5beaa1208c1d0271bbac91825f56478d483d", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "989f36d47217386898acefac9f60881d55b0333d9e39df448e0384e7887db8b7", "generation_seconds_batch": 17.727117046015337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.078212261199951, "genuineness": 4.193545341491699, "identity": 0.6922745704650879, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d8f60f39c6c3b88a2b5e622ce4ff5beaa1208c1d0271bbac91825f56478d483d.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "d9159b04cc01d96d24f0c6d2c117bcbcfcc5c33a5263be4d8389900f56a910ab", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4a7c3bb548205d294eaf2f37153fb0327cb065c682151c653e29747daf32ef03", "generation_seconds_batch": 23.265318181016482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.739579677581787, "target_z": 0.8437500731991853, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.25392571091651917, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d9159b04cc01d96d24f0c6d2c117bcbcfcc5c33a5263be4d8389900f56a910ab.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "d9394e511b87a59953cd81be2bdf50a4e538ae0be86e58135c6be8c05f689b3b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c4b93e876165ae693d78cb307c4197428d7040fba455097b65af91dbcab68e12", "generation_seconds_batch": 24.628098882036284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.524835109710693, "target_z": 1.460937494876057, "ce": 6.006728172302246, "genuineness": 4.557555198669434, "identity": 0.7250628471374512, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d9394e511b87a59953cd81be2bdf50a4e538ae0be86e58135c6be8c05f689b3b.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "d9650eacdc2a0995695777b5997127fd5918532849d4d6f40dae5b633648a7b1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "98924d843866b61af921a23ca28f4338739c47ac186b8da452d03b0c695913f6", "generation_seconds_batch": 23.305688432999887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 6.252951145172119, "genuineness": 4.15588903427124, "identity": 0.4469031095504761, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d9650eacdc2a0995695777b5997127fd5918532849d4d6f40dae5b633648a7b1.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "d9c7b3c0ab714059ef5fae1429c885c11b8e1425e420aa0e215cf0e722ab0c57", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c6cbce343bde7bd0fcaacf47ed08a346afde1b5d30a41701242a24a9042823da", "generation_seconds_batch": 17.107972488971427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8837087154388428, "target_z": 0.9570312115704277, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.6033328771591187, "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_S_MONO_reg/d9c7b3c0ab714059ef5fae1429c885c11b8e1425e420aa0e215cf0e722ab0c57.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "d9e82decd3947198886277d101d3f0e9e229c696f5a4dd55751248dfa879baa6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e96488e7669f0f49974c7eaf9ddb6f73ab8e89682eb5d047006be236edd0c575", "generation_seconds_batch": 22.783045891905203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.077537536621094, "target_z": 1.1093749019130918, "ce": 6.125868320465088, "genuineness": 4.557555198669434, "identity": 0.6247981786727905, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/d9e82decd3947198886277d101d3f0e9e229c696f5a4dd55751248dfa879baa6.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "da0ebddb7a46542c8a1621a81d13124923ea355ab92eb0a461fdbcc0122ce033", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8b85f8f8bc2b7276a6927a5443863559198a14ae0a24d4c2f6f2feb2d936770b", "generation_seconds_batch": 16.72134528902825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3718016147613525, "target_z": 0.5546874509565459, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.5721925497055054, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/da0ebddb7a46542c8a1621a81d13124923ea355ab92eb0a461fdbcc0122ce033.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "da1421f25751e924910bfdd648852b39f17ae438a9bc876ee92b02e44c843e8c", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2e65fa788d494acad552ae28d16cc1b6cdbc9ca4a831b928fc52979865f2bfd", "generation_seconds_batch": 27.372245289036073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.09409761428833, "genuineness": 3.9801597595214844, "identity": 0.6683290004730225, "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_S_MONO_reg/da1421f25751e924910bfdd648852b39f17ae438a9bc876ee92b02e44c843e8c.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "da566db9a13f5124df2914629cdc05c9dfb2b7cffbdbb4a8ed26932856badb45", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "975cd45428a344ba0564bce7f4ead66b0e7070e39b364b3e3746db5827b6cdaa", "generation_seconds_batch": 27.785415231017396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.260893821716309, "genuineness": 4.369274139404297, "identity": 0.7926665544509888, "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_S_MONO_reg/da566db9a13f5124df2914629cdc05c9dfb2b7cffbdbb4a8ed26932856badb45.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "18dfc67330a98803991b22aebfadd42090c28e320fee652119d9f87da6d5fb72", "generation_seconds_batch": 18.197057637036778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823862314224243, "target_z": 0.12402345259733197, "ce": 6.300607204437256, "genuineness": 3.9927117824554443, "identity": 0.47805702686309814, "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_S_MONO_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "db24bd13a8d7819ff6ae0f67231f718e0debd435a691f0878986ab6ae6b5802c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6802564cdfa07c33ea41b7609da8d0f02029c2a3db4f708da6c135b26bccf642", "generation_seconds_batch": 20.275241427938454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 6.1894097328186035, "genuineness": 4.720732688903809, "identity": 0.7866395711898804, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/db24bd13a8d7819ff6ae0f67231f718e0debd435a691f0878986ab6ae6b5802c.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "db615fd15b961e7a9189e2a70b21ffee99cc4b5da632ebd219e38b4d4bd8e1f9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8fab567352358f1899c539b8dc98aabc2ee2f8d5a0fa40c6ae00d44b86a50f04", "generation_seconds_batch": 27.221027806052007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2077927589416504, "target_z": 0.4257813667527005, "ce": 6.403861999511719, "genuineness": 4.375550270080566, "identity": 0.4525240659713745, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/db615fd15b961e7a9189e2a70b21ffee99cc4b5da632ebd219e38b4d4bd8e1f9.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "db9fa994828a218f6d24c94ab9ccc4a16d2491b22d1628cb14d18d0c8c92aac8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4da47ba2789b5ce865d12701eb083cfe947870172c73f1f0116009baf81a217f", "generation_seconds_batch": 22.673166669905186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.186877250671387, "target_z": 1.1953126661621505, "ce": 5.8875885009765625, "genuineness": 4.20609712600708, "identity": 0.6057242155075073, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/db9fa994828a218f6d24c94ab9ccc4a16d2491b22d1628cb14d18d0c8c92aac8.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "dbbc10751ed81f90be6afbac953cf3191bc0ea964ac6105893fddcda90732ec7", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "824bc68801407df06b7faa1f3b7dddd91338e02b5c0aab5853c2efb816711eb0", "generation_seconds_batch": 17.52979721489828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 6.109982967376709, "genuineness": 4.3127899169921875, "identity": 0.720302939414978, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dbbc10751ed81f90be6afbac953cf3191bc0ea964ac6105893fddcda90732ec7.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "dc2c0a85263a55ad40e82dd11a5374798fb43afade8a48690ea86dcbb7cb07cc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3cefa4584ec0f018e7c7a83eba52f757e42dd55a652995177ac6c6e0acc81fee", "generation_seconds_batch": 24.105277594993822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 6.3641486167907715, "genuineness": 4.293961524963379, "identity": 0.5332711935043335, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dc2c0a85263a55ad40e82dd11a5374798fb43afade8a48690ea86dcbb7cb07cc.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "dc2ede064439661126137ce0a415cdccdc4645b21369d81b66503fb67ce133de", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e543e4ad53efcd9db5a84427d5c6132ecf5eb3a7409feced9e41aae4ad70d082", "generation_seconds_batch": 18.58606014493853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.739579677581787, "target_z": 0.8437500731991853, "ce": 5.9431867599487305, "genuineness": 3.9801597595214844, "identity": 0.5243725776672363, "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_S_MONO_reg/dc2ede064439661126137ce0a415cdccdc4645b21369d81b66503fb67ce133de.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "dc9d018349ff2d9c0afdd023484128c28b1192f220175d6c504680033a5aecdd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7cb58316561d3af1e4e216ae22fd9b9a52ff431661434b8ea06aecb4d77556bd", "generation_seconds_batch": 23.51376200595405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9483184814453125, "target_z": 1.0078125666112585, "ce": 5.839932441711426, "genuineness": 4.545003414154053, "identity": 0.7594974637031555, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dc9d018349ff2d9c0afdd023484128c28b1192f220175d6c504680033a5aecdd.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "dcf25988ed93496f19a56bbbb82b2e7aef4cae24917a89a882ab4e0b4b5b5c80", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "15bb08361d12099e974938ff1f3c58099f3ea24ed63129eb2ebdc800e15d55a5", "generation_seconds_batch": 22.7423298220383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.892612934112549, "target_z": 1.749999929728782, "ce": 6.038498878479004, "genuineness": 4.68307638168335, "identity": 0.8915899991989136, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dcf25988ed93496f19a56bbbb82b2e7aef4cae24917a89a882ab4e0b4b5b5c80.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "dd0873225e1e1aff801b5ac7ce0350a0615eceb6dc68e85018a6896c6ee9b84c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0c7a112389e8b4e6d57e47fb22e67534afb7c9ca5f416126b3bd8fa96ca1109f", "generation_seconds_batch": 26.52677231898997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 5.982900142669678, "genuineness": 4.444586753845215, "identity": 0.5057520866394043, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dd0873225e1e1aff801b5ac7ce0350a0615eceb6dc68e85018a6896c6ee9b84c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "dd69198bbc5bbb64a6eef6d39aa96fa28c926279618e8fb7b14f1f6e9a56cb90", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "741da2237a325a7bb64e57e19ea44076d346ea93b770fb941482d8f921898734", "generation_seconds_batch": 18.03129886800889, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6501200199127197, "target_z": 0.7734374421726437, "ce": 5.927301406860352, "genuineness": 4.193545341491699, "identity": 0.6738200783729553, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/dd69198bbc5bbb64a6eef6d39aa96fa28c926279618e8fb7b14f1f6e9a56cb90.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "ddb30af4f9fca36016da9a9ad3ff7e25186214d414b28a7f919baf1a4568f225", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4d95343288ea966ddf9d16b567cea9ae36dcc00ffd814a605ef17c8b0797afe9", "generation_seconds_batch": 23.129189759027213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316096305847168, "target_z": 1.2968750014639836, "ce": 5.998785495758057, "genuineness": 4.595211505889893, "identity": 0.6314026117324829, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ddb30af4f9fca36016da9a9ad3ff7e25186214d414b28a7f919baf1a4568f225.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "ddf68da1a8b0a2db8db5226db60042e61e5d8cc8f787e77802f1fd57ac890f45", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4ae9100625b45c225fe45720ae5a24f6005bc53d86366861d6ebc459d519fcd6", "generation_seconds_batch": 17.022644135984592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.1020402908325195, "genuineness": 4.532451152801514, "identity": 0.6414942145347595, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ddf68da1a8b0a2db8db5226db60042e61e5d8cc8f787e77802f1fd57ac890f45.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "de2867a235feaf858891b0b613a9ec538b4e0e6d42b88b11313d1feaaf4f4c05", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c491f219b6ee567de374d7536e7b4b15d8631c53db6507a6849220f2e22f088a", "generation_seconds_batch": 22.293837398057804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.763393402099609, "target_z": 1.6484372196471204, "ce": 5.700935363769531, "genuineness": 4.6956281661987305, "identity": 0.726550817489624, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/de2867a235feaf858891b0b613a9ec538b4e0e6d42b88b11313d1feaaf4f4c05.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "dfa3a047c6fcf890b3054584058d09d77d024784ab52c75b4d6c3f688acfbdd6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cabd5291d8b8fbf4dbdf7e10cdd87048659a74893e71785dddcb81df14adcc54", "generation_seconds_batch": 25.31284831801895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.250037431716919, "target_z": 0.4589844234944602, "ce": 6.213237762451172, "genuineness": 4.287685871124268, "identity": 0.3629545271396637, "hypothesis": "I found the little box. It was beside the old window. We can look inside to", "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_S_MONO_reg/dfa3a047c6fcf890b3054584058d09d77d024784ab52c75b4d6c3f688acfbdd6.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "e017e9ebbc8b8e82261ce334f0c33c66c6d6680b919cb5d1c973f14881743295", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "05389a37215f6413d0daeb676cef85f1918ec2d303175ff276293f9df5533c6a", "generation_seconds_batch": 17.02790195296984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4364113807678223, "target_z": 0.6054688059973767, "ce": 6.1894097328186035, "genuineness": 4.243753433227539, "identity": 0.7109955549240112, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e017e9ebbc8b8e82261ce334f0c33c66c6d6680b919cb5d1c973f14881743295.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "e06479a63f37d04a8f3965a65693aaba6c0a931c209eaac7fd2527e8ea2aa96b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4d00358a5c8d8a6ddc78443e4fd6b6fe00eba9587e93523b0928e17716a9dcfd", "generation_seconds_batch": 15.853977181017399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.785966157913208, "target_z": 0.09423820050214875, "ce": 6.213237762451172, "genuineness": 4.570107460021973, "identity": 0.5063416361808777, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e06479a63f37d04a8f3965a65693aaba6c0a931c209eaac7fd2527e8ea2aa96b.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "e08939bdd0461a0f6de503fbdf4e537bea1d025e6983e5c54969cf39075df117", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74057fc98eae6db80283c92a86bffe8daa0b04408d8a2cedd28486f2714d866a", "generation_seconds_batch": 16.890796623076312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282730102539062, "target_z": 0.36328120864246033, "ce": 6.348263263702393, "genuineness": 4.7081804275512695, "identity": 0.5398979187011719, "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_S_MONO_reg/e08939bdd0461a0f6de503fbdf4e537bea1d025e6983e5c54969cf39075df117.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "e12a04507e5c832d92b4ca3300c497b1639b595fdd3311ae91fd9931a8d7a164", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cf2053cea4b7ac936165259002ea11623ab47516f6d6de85c6fef8a2bac962ec", "generation_seconds_batch": 20.62212675693445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3320419788360596, "target_z": 0.5234375592913401, "ce": 6.38797664642334, "genuineness": 4.444586753845215, "identity": 0.41569018363952637, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e12a04507e5c832d92b4ca3300c497b1639b595fdd3311ae91fd9931a8d7a164.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "e12e610dec32afcdea1547a57a4f4ae36b0ca472bc53d0182b161a7e24c63b54", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9920503d09b27cd4cf92fab3bb88a5737cd56ca0351a407c53b767f32bcb950b", "generation_seconds_batch": 16.978754018084146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.117297172546387, "target_z": 1.1406247935782976, "ce": 6.1894097328186035, "genuineness": 4.444586753845215, "identity": 0.7894538044929504, "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_S_MONO_reg/e12e610dec32afcdea1547a57a4f4ae36b0ca472bc53d0182b161a7e24c63b54.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "e17d3f4c86f4cd2536c0dbbd6c03b4239b50fe74c3c837170286ec54022d888d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1ec419cdfb659e57f4bed1d7f0b0bc69f5eb389433a37deaba7f47a0c3da08d3", "generation_seconds_batch": 19.76527239906136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.077537536621094, "target_z": 1.1093749019130918, "ce": 6.2767791748046875, "genuineness": 4.532451152801514, "identity": 0.7391701340675354, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e17d3f4c86f4cd2536c0dbbd6c03b4239b50fe74c3c837170286ec54022d888d.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "e1e9f14fbf4328c3f79a64942ec1422c8342fde2ea1d32773157bf79b9bb65f6", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3aa2c2e5770aea22d06676f3e2de0af933a14a2fd3a21efc33e634f633507f44", "generation_seconds_batch": 29.08080533402972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.981658935546875, "target_z": 0.24804690519466394, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.7284409999847412, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e1e9f14fbf4328c3f79a64942ec1422c8342fde2ea1d32773157bf79b9bb65f6.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "e262a6d7d1271afdb96cace8a4b145506d5930a040f4be9efbae2193b254529b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "9ef3b8bdce0bb6715f8bd718fd1e60633b1dfef3f588f31fb1ddab71896a6da0", "generation_seconds_batch": 22.859746392001398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6103603839874268, "target_z": 0.7421875505074378, "ce": 6.324435234069824, "genuineness": 4.469690799713135, "identity": 0.6659194827079773, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e262a6d7d1271afdb96cace8a4b145506d5930a040f4be9efbae2193b254529b.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "e2a50fd648e2716de2b99427579b2a8cf6694ab441bc386c25b5936a3131bf54", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "df9f6a2e2cb184c59e2c113f509a27eabef9fe1f94565044d4543e3d528f4998", "generation_seconds_batch": 15.573615973000415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8164074420928955, "target_z": 0.11816414475758444, "ce": 6.022613525390625, "genuineness": 4.250029563903809, "identity": 0.6496893167495728, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e2a50fd648e2716de2b99427579b2a8cf6694ab441bc386c25b5936a3131bf54.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "e3175abc2c894c33f20176773c9d7844cd7b6ed8e542ca90d50563ebef584fcd", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "773490e391d05d015e34a718c19d28db746f3ab7ad277ad84a340a704d6c45a6", "generation_seconds_batch": 22.48715153406374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7147297859191895, "target_z": 0.8242187972134745, "ce": 6.141753673553467, "genuineness": 4.049196243286133, "identity": 0.24202625453472137, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e3175abc2c894c33f20176773c9d7844cd7b6ed8e542ca90d50563ebef584fcd.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "e39d85f9bd86d93f2a9ef133a4a5b9522dbb2907d091190aee6f5cdabbeb8482", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69e04326ee7a18abb8502fd3e88454620f73efedf35486bd5e91f8bef1a0b76c", "generation_seconds_batch": 27.77285269391723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 5.601652145385742, "genuineness": 4.432034492492676, "identity": 0.6974791288375854, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich lass sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e39d85f9bd86d93f2a9ef133a4a5b9522dbb2907d091190aee6f5cdabbeb8482.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "e3af5cf8706cfba89dd2b8e88391a63fd627442db6c4c3b51f6d805125ba627c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "612c3b4f2916e140ac7c94960d962a565691a702e1ac9daabacd697b0632109f", "generation_seconds_batch": 16.230958738015033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.550720691680908, "target_z": 0.6953125256197148, "ce": 5.855817794799805, "genuineness": 4.356721878051758, "identity": 0.7537243366241455, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e3af5cf8706cfba89dd2b8e88391a63fd627442db6c4c3b51f6d805125ba627c.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "e41c8eeb6e7ccacd37ff536fcf658895e74875cda8de2a256681084a724b7faf", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d305e0968f7d560b610456db65db5995950e5c87c87ca4afcedb7365eb6bb90b", "generation_seconds_batch": 23.20635403401684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.62030029296875, "target_z": 0.7500000234237393, "ce": 6.252951145172119, "genuineness": 4.469690799713135, "identity": 0.5530902147293091, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e41c8eeb6e7ccacd37ff536fcf658895e74875cda8de2a256681084a724b7faf.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "e4490effc1fcc46aa6c4f0d2a34dcaef34027746562e7a62b04a3e28dd462c8e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8ed88d4f25a02f44b79812a4da22af2d0ec9b34995ef60a1a0ba5e070262c934", "generation_seconds_batch": 22.943371178931557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5703938007354736, "target_z": -0.07519526263305504, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.6475805044174194, "hypothesis": "De deur 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_S_MONO_reg/e4490effc1fcc46aa6c4f0d2a34dcaef34027746562e7a62b04a3e28dd462c8e.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "e44c66b935199b845d62f64a4ef6cca7e7e3940a546523dae8a9183bd1df555a", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "bec6ffd141c0e63a7db59bfc45e8a5d576d5fec451c5fc2aab95fcb52a9eff9f", "generation_seconds_batch": 19.996973425964825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9841439723968506, "target_z": 0.25000007027121784, "ce": 6.1894097328186035, "genuineness": 3.7667744159698486, "identity": 0.2856425642967224, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e44c66b935199b845d62f64a4ef6cca7e7e3940a546523dae8a9183bd1df555a.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "e4666ea04636609dd5d76b7dd087fe01780443ae261df5181449eff03a36fe49", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "df0c46da4bd63695f2f9b58d95475276b6d1f9829c6f0dc0274929cf78551cf3", "generation_seconds_batch": 17.133255761000328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.277372360229492, "target_z": 0.48046886455672494, "ce": 6.395919322967529, "genuineness": 4.645420074462891, "identity": 0.4412488639354706, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e4666ea04636609dd5d76b7dd087fe01780443ae261df5181449eff03a36fe49.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "e4db406e7d9c977b9e1d9aaa83df77ed66eb517a49c84669598fe48481a3eb00", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "296f235d0e575c2c53ce246b6af2ecffb922c2fa16171dd0a3103ce555936eb1", "generation_seconds_batch": 28.642807152005844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.819099187850952, "target_z": 0.9062500439195111, "ce": 6.419747352600098, "genuineness": 4.256305694580078, "identity": 0.7577389478683472, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e4db406e7d9c977b9e1d9aaa83df77ed66eb517a49c84669598fe48481a3eb00.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "e51cb80843feb66f993eb3096f65d86ac6378307672999abbe6400903184eba2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "229d47f9a77604112060c9eb498840718127fd7dbb1395d0d7d7437d10fc5a84", "generation_seconds_batch": 22.675085710012354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.06759786605835, "target_z": 1.1015626163867045, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.3890552520751953, "hypothesis": "De deur 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_S_MONO_reg/e51cb80843feb66f993eb3096f65d86ac6378307672999abbe6400903184eba2.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "e56fab285a176c9a0056b533a407658982850b024ab64fc82112721c764e3a3e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bf3322cbefe638fdb30d9c98c9dce12c655a903ab84fb01858f3d8ed6f9660fa", "generation_seconds_batch": 22.640005494002253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.395919322967529, "genuineness": 4.482243061065674, "identity": 0.6476345062255859, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e56fab285a176c9a0056b533a407658982850b024ab64fc82112721c764e3a3e.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "e5efa52f7c54428ddb9d7fb966f68a6149bdb56efb461927fabcaac76e96ee05", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "de660200b98764bb360de221203941f0eb861fb3bd17250b98bd458dbec7263e", "generation_seconds_batch": 17.17114224296529, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1506378650665283, "target_z": 0.38085931955161717, "ce": 6.181467056274414, "genuineness": 3.9425034523010254, "identity": 0.41640138626098633, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e5efa52f7c54428ddb9d7fb966f68a6149bdb56efb461927fabcaac76e96ee05.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "e5f4570e5d66379626ee7bc064eb38283f5354898d095dd3850997dead5a6ec0", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "3434037d677e73a0263042764163f1de21bd770ced5073d83d3e295c556e2068", "generation_seconds_batch": 18.86498992203269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.505990982055664, "target_z": 0.6601563038014012, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.7340022921562195, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e5f4570e5d66379626ee7bc064eb38283f5354898d095dd3850997dead5a6ec0.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "e5fad2b2ba16c5178f6186fde4a9b8f33c58d8673f516af4f0632c57003a1939", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7359a868ea9f49978f60a4143dee204c35b22fc4bcf72ab6a097b1f0080fb801", "generation_seconds_batch": 23.183315658941865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2127625942230225, "target_z": 0.4296875095158941, "ce": 6.078212261199951, "genuineness": 4.3127899169921875, "identity": 0.7641674876213074, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e5fad2b2ba16c5178f6186fde4a9b8f33c58d8673f516af4f0632c57003a1939.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "e64152ebc088b76a984827a856da0139c9cdaa650eaecf13f752b0d7e3ec184f", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "63e42d9425327fad494eea31bc55a44e237c7420bcc44a116f407ab2ae5832f8", "generation_seconds_batch": 16.903253169031814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.673933982849121, "target_z": 1.578124776010493, "ce": 6.260893821716309, "genuineness": 4.65797233581543, "identity": 0.8101379871368408, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e64152ebc088b76a984827a856da0139c9cdaa650eaecf13f752b0d7e3ec184f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "e67fce480bd228772621a0b06bcac09dfb283b2d9466778af590b42b6b4a3cc8", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7c22dd8666fbb6889b525501ae1306e8c6b5f8b5e6e9a2cce9e3ec8ce08a9de6", "generation_seconds_batch": 23.23275757499505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.216696739196777, "target_z": 1.2187498975211406, "ce": 6.308549880981445, "genuineness": 4.362998008728027, "identity": 0.8136130571365356, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e67fce480bd228772621a0b06bcac09dfb283b2d9466778af590b42b6b4a3cc8.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "8ba4b07f873365c75e35b0dac49310d4136efa6ba6f7f6e6049258a293274024", "generation_seconds_batch": 14.967284665093757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7126591205596924, "target_z": 0.03662104877825055, "ce": 6.3641486167907715, "genuineness": 4.582659721374512, "identity": 0.28317856788635254, "hypothesis": "Someone moved the little blue chair. I used to stand here beside the tall bookshelf. We can put it back after lunch?", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "e6f50681a0a851ca9d340ec78a7299db2ffc4e2351ed12b92d00edcc65b198e7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "89e249e12eee7a9f3803624fb47256a0a66cd63d0a8227b45c48255caf0c2e57", "generation_seconds_batch": 26.707850712933578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.38567590713501, "target_z": 1.3515624992680082, "ce": 6.038498878479004, "genuineness": 4.582659721374512, "identity": 0.730003297328949, "hypothesis": "La porte 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_S_MONO_reg/e6f50681a0a851ca9d340ec78a7299db2ffc4e2351ed12b92d00edcc65b198e7.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "e7203151f659ef81c017fc4d686359ed10d8fc03a391b044cbfb75adc29b675d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "7f46ed3f846188ad7446e895ef37206de4556edebb5ce9545c88e2a286629c78", "generation_seconds_batch": 26.707850712933578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.6925017833709717, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e7203151f659ef81c017fc4d686359ed10d8fc03a391b044cbfb75adc29b675d.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "e7a7e36dc253b2c5fe0e0cfec1798b64515b0044fb3e324ca9f0c83465d30837", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "814023c5872c0284dda2d128fa4a3c856437dee2e064c4cdeee4a6213d46b264", "generation_seconds_batch": 25.259155779029243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25645637512207, "target_z": 1.2499997891863464, "ce": 6.133810997009277, "genuineness": 4.570107460021973, "identity": 0.5827003717422485, "hypothesis": "La porte 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_S_MONO_reg/e7a7e36dc253b2c5fe0e0cfec1798b64515b0044fb3e324ca9f0c83465d30837.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "e814f9b19669596e239f178155332dfff5116b7fb3becd2f42c3cc314bf51132", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a0c43830008ab4d0d94fb21ab375a57607ced3b27e88d85aff38b7a8a2585b47", "generation_seconds_batch": 17.924006558954716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3320419788360596, "target_z": 0.5234375592913401, "ce": 6.173524379730225, "genuineness": 4.494794845581055, "identity": 0.703501284122467, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e814f9b19669596e239f178155332dfff5116b7fb3becd2f42c3cc314bf51132.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "e88711b20d6b277960ef86aa0b26c6ca3b96632a8bf39c75349797350aec6f2f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c7a0db2cec46e167e30033561a36b23312e8f9ae787d428051a57a5c6fa53a33", "generation_seconds_batch": 22.397945323958993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.165581703186035, "genuineness": 4.049196243286133, "identity": 0.3398425579071045, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e88711b20d6b277960ef86aa0b26c6ca3b96632a8bf39c75349797350aec6f2f.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "e8a686730240ec8d61d6d880c25db516a5942872f33dc9fad70e4b50b0273f8c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "743b0fbef6fa250a0a30db5f4b39c8ff3934784be468c5b6bebe51357b280215", "generation_seconds_batch": 16.706398878013715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9354798793792725, "target_z": -0.5742187269422566, "ce": 6.141753673553467, "genuineness": 4.256305694580078, "identity": 0.4545901119709015, "hypothesis": "De deur staat nog open. Ik heb je Jazi op de stoel gelegd. Wanneer je klaar bent.", "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_S_MONO_reg/e8a686730240ec8d61d6d880c25db516a5942872f33dc9fad70e4b50b0273f8c.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "e91cac5f51015d1894eae23f8a45f6426e564db6356acdf87c43c51bea7065c9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4f39b5eb29bcc638f7ae1146014abce9ed3c5e5c2f6d902106ea8053a7c70098", "generation_seconds_batch": 17.850500177009962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.857020378112793, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e91cac5f51015d1894eae23f8a45f6426e564db6356acdf87c43c51bea7065c9.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "e9246ebed7d5603a953ac085201f4be731a3b0fc95bde0e920081e940d3c968c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91080f65b21c92ca1b081569e9883c911df015c963befc790b4f618fbfa96a52", "generation_seconds_batch": 25.010454149916768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.127237319946289, "target_z": 1.1484374538845132, "ce": 5.935244083404541, "genuineness": 4.011539936065674, "identity": 0.43601271510124207, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e9246ebed7d5603a953ac085201f4be731a3b0fc95bde0e920081e940d3c968c.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "e93b1b7c53aebd4cee9db9094f8d2b73b48e096cc2933a420a3ea7a7b8d8e1f4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "dd4f1bbb3455c6f70f15ddf50613a43a0bc2ab57e9cb325b475211e48403f669", "generation_seconds_batch": 25.102356048999354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6352100372314453, "target_z": 0.7617186391032343, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.7108193039894104, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e93b1b7c53aebd4cee9db9094f8d2b73b48e096cc2933a420a3ea7a7b8d8e1f4.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "e97bfd12861c5a0b3bf317eefb4f233e522c96c19971b23d233432574d56df4b", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12cdcef0aa7692c60d437b4cbed7c98d1aff0b5f686863f6b247cc18bd9d8c8d", "generation_seconds_batch": 27.43196982401423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.819099187850952, "target_z": 0.9062500439195111, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.663117527961731, "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_S_MONO_reg/e97bfd12861c5a0b3bf317eefb4f233e522c96c19971b23d233432574d56df4b.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "e99d7dcd01568644232e840128646bc845f6ba008fa080b3ee43229995d91f17", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a92f692dd8c6f9f599006ad2804189d3364b7f07e3ae76ec37523dc048bbac53", "generation_seconds_batch": 23.697775250067934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.332377910614014, "genuineness": 4.519899368286133, "identity": 0.3670806586742401, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/e99d7dcd01568644232e840128646bc845f6ba008fa080b3ee43229995d91f17.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "3bf4d301c3fa35a9f9617d87397474be3394b7e92b2eacb489e42e72c7b0c0e8", "generation_seconds_batch": 16.28890575398691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192882776260376, "target_z": 0.4140625636832912, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.5682026147842407, "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_S_MONO_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "ea1221b77c21079d12d2c267c035b2bac6930bcb82917c41d2849d78fd759b3d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "36ee2f3fa0598333355697e70d6968490b9ae38a494f614e029afe805c8b3e51", "generation_seconds_batch": 22.944575543981045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.38003396987915, "genuineness": 4.607763767242432, "identity": 0.5722087025642395, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ea1221b77c21079d12d2c267c035b2bac6930bcb82917c41d2849d78fd759b3d.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "ea3a85bb83e956c776e899faee5cd900d47e63b3d59fde0a0d4744bf8f8c71ef", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "60cd16fda2a4a72f44215af760348feac17c303e90bef43ff0ec23bcb0a57660", "generation_seconds_batch": 22.859500022022985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689879894256592, "target_z": 0.8046875212277638, "ce": 6.165581703186035, "genuineness": 4.632867813110352, "identity": 0.6582908034324646, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ea3a85bb83e956c776e899faee5cd900d47e63b3d59fde0a0d4744bf8f8c71ef.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "ea5f5f4026895b84c4362a675682fa58658d6f65ff1dc00643e1e2f5138b318f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "84e5d740a87d818812e7f9b4fc1f2e50700383ec66703d413bf136f7b8366089", "generation_seconds_batch": 22.397945323958993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.1020402908325195, "genuineness": 4.080576419830322, "identity": 0.41277724504470825, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ea5f5f4026895b84c4362a675682fa58658d6f65ff1dc00643e1e2f5138b318f.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "eaa11096197f2af4c8fe41ae3fb5f2689ae9177be94ec56bed832cb45326431b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e66df6acd1ba8d855cb071434d7262315dedb5b2470a8343fe0d33ca91db09d9", "generation_seconds_batch": 23.771782980998978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 5.7247633934021, "genuineness": 3.9801597595214844, "identity": 0.5336369276046753, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/eaa11096197f2af4c8fe41ae3fb5f2689ae9177be94ec56bed832cb45326431b.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "eb0897d899718116714178475ef6ebd2aea619371fabc501ed36b83789230435", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "71c2ea22319b485cc85acf925527d726d7ac3ee6296d5895c5c84f163e307ff8", "generation_seconds_batch": 17.693502653972246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5943117141723633, "target_z": -0.05639649382226985, "ce": 6.205295085906982, "genuineness": 4.337893962860107, "identity": 0.6948953866958618, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/eb0897d899718116714178475ef6ebd2aea619371fabc501ed36b83789230435.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "eb36a747457f70e666f11fb663fb1d3c13d18ef8ba559daf7ae8406055098d41", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1aa4d0257443e07c05798cbbb686a68b7976ac02ff261a826df2470972d76bd8", "generation_seconds_batch": 17.376624096999876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.703754425048828, "target_z": 1.6015627569291402, "ce": 5.33954381942749, "genuineness": 4.093128204345703, "identity": 0.7684070467948914, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/eb36a747457f70e666f11fb663fb1d3c13d18ef8ba559daf7ae8406055098d41.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "eb56668dbd9c84e784e6e3342becbcfb854b7dba0938fe0359da75d3e9bffa3c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d467b4feee531d00d01fa2dbe88c9d92efefb4050109efa5957177a9effc7afb", "generation_seconds_batch": 16.64903970900923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.165581703186035, "genuineness": 4.13078498840332, "identity": 0.6709855198860168, "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_S_MONO_reg/eb56668dbd9c84e784e6e3342becbcfb854b7dba0938fe0359da75d3e9bffa3c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "eb98c5f7ac153a66ab53e9b96929402371de5f3a5180df4b42560e37d1f02fa0", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "30c65bb1619ccf2f813429b8f1913b7be92a05f8d5633e0a1c084662927d348f", "generation_seconds_batch": 16.526820008992217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.417567014694214, "target_z": -0.1953125724671934, "ce": 6.0305562019348145, "genuineness": 4.168440818786621, "identity": 0.613828718662262, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/eb98c5f7ac153a66ab53e9b96929402371de5f3a5180df4b42560e37d1f02fa0.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "6abb076da06a0bd34202e89e85903ad1ff0c8374035fbb655ed7c6c81a03252e", "generation_seconds_batch": 15.186213168082759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6849098205566406, "target_z": 0.8007811910746558, "ce": 6.149696350097656, "genuineness": 4.394378662109375, "identity": 0.7758889198303223, "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_S_MONO_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "ec13074d7d27986eb73f4d88ff32341144e9d9aa5c886b21ff0e1aa4e867c7e4", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8c14a84748ca1194e2fcd83d6a4e8cb86ad3638bf9fdb84374b138536982fc7d", "generation_seconds_batch": 22.768882901058532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.176937103271484, "target_z": 1.1875000058559348, "ce": 6.141753673553467, "genuineness": 4.319066047668457, "identity": 0.6107636094093323, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ec13074d7d27986eb73f4d88ff32341144e9d9aa5c886b21ff0e1aa4e867c7e4.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "ec62af8348b4f6772584eec96970ae8ce2c7e257bb8b4b137132824befa005eb", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "829ee0d2c83314964948573bd7bdec4bfdb9247bd8ee2041a5ee7a036e4afe84", "generation_seconds_batch": 17.660411798977293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6401801109313965, "target_z": 0.7656249692563422, "ce": 5.847874641418457, "genuineness": 4.005263805389404, "identity": 0.527588963508606, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás linda.", "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_S_MONO_reg/ec62af8348b4f6772584eec96970ae8ce2c7e257bb8b4b137132824befa005eb.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "ec70fb9fc342a6f0a8f89b3cbf587fe3312e2e1e086716719c9735045e24160f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cbb5c20826e33267b459cce4bd843f4ced38da5cadd05fdb525c804da7e4a30f", "generation_seconds_batch": 17.50185189000331, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5258708000183105, "target_z": 0.6757812496340041, "ce": 6.356205940246582, "genuineness": 4.519899368286133, "identity": 0.6213542819023132, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ec70fb9fc342a6f0a8f89b3cbf587fe3312e2e1e086716719c9735045e24160f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "ecc6d1a3399cd65a498b67c8e625bfe31e3981afc05d4b486d1dae489701990d", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "190aa990f828367fa14fb68e654915741d2627cb462492cf6326c80c1b0fd1fb", "generation_seconds_batch": 17.37892849103082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5556907653808594, "target_z": 0.6992188557728227, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.490177184343338, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ecc6d1a3399cd65a498b67c8e625bfe31e3981afc05d4b486d1dae489701990d.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "ed9d565f55ca4280bd9c9f1a72558c9b6a1d8040878ef859d5cfd835ee756410", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "8d101c9eb8b9ab7d9b3c841d19df7b86d6122770078e04e95502fbf77eb89c3f", "generation_seconds_batch": 28.642807152005844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.5055819153785706, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ed9d565f55ca4280bd9c9f1a72558c9b6a1d8040878ef859d5cfd835ee756410.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "edbdc02f2caeb93ac5e243088ab2994d57233e2326956b35625d26c177b8f829", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e9cbc74b071247b4afbbd44b9877ba50e84cdd7a8ee0c459a310bf12931422b", "generation_seconds_batch": 16.96097723499406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4950151443481445, "target_z": 1.4374998887372383, "ce": 5.95112943649292, "genuineness": 4.758388519287109, "identity": 0.8393001556396484, "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_S_MONO_reg/edbdc02f2caeb93ac5e243088ab2994d57233e2326956b35625d26c177b8f829.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "ee550681e4f19c5477f62f21d65bea2d5e2d0c1327f62d1abbe256beaf2a60c3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0421784f8c6dfb5ebf9cf50758ee13c51e8f14b93c95d371eca6e35a7ead23f3", "generation_seconds_batch": 24.389690459007397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5706005096435547, "target_z": 0.7109374714523178, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.46837785840034485, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ee550681e4f19c5477f62f21d65bea2d5e2d0c1327f62d1abbe256beaf2a60c3.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "eea23ced1247dc7df5aedd191688e686a7d1851c528bb385f6920394577610c1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0acec814d70d10601f44faef5a304440825fd31e00fe053e79daca02a44bcd91", "generation_seconds_batch": 23.794424440013245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.395615577697754, "target_z": 1.3593747847943953, "ce": 6.165581703186035, "genuineness": 4.921566009521484, "identity": 0.6597849130630493, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/eea23ced1247dc7df5aedd191688e686a7d1851c528bb385f6920394577610c1.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "ef079eb9f84ad45f6a824e768f23dfe5475c58ada43a0fe435e4142281e1d0a9", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ac2d950967badc51278569aa04420ce1c935b321230a7447987b1824d8bbde67", "generation_seconds_batch": 22.34321841993369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.046441555023193, "genuineness": 4.250029563903809, "identity": 0.7708333730697632, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ef079eb9f84ad45f6a824e768f23dfe5475c58ada43a0fe435e4142281e1d0a9.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "f01b2215c9cea3d0be3f986ccee5fe1e562978f1805ea34d61a04afecb0c5bd7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a20efc198c9d53b1933adb9764b211cbb6b9b3291c2f172f8f574f53fb8b3729", "generation_seconds_batch": 23.886859060032293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4115614891052246, "target_z": 0.5859375300116659, "ce": 6.197352409362793, "genuineness": 4.137060642242432, "identity": 0.690382719039917, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f01b2215c9cea3d0be3f986ccee5fe1e562978f1805ea34d61a04afecb0c5bd7.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "f053cd63d28c73a33f2151ea1cd1fe19e25c0bef8f902357c4496a9a10f7aab0", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "6c863ae4f459028a50a793045facf9ee90138227ab595358409348a3422ad8b5", "generation_seconds_batch": 27.92381970596034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246516704559326, "target_z": 1.2421875036599592, "ce": 6.133810997009277, "genuineness": 4.444586753845215, "identity": 0.8447949886322021, "hypothesis": "Colcan a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.047619047619047616, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f053cd63d28c73a33f2151ea1cd1fe19e25c0bef8f902357c4496a9a10f7aab0.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "f06e3c4906bcd4f0925c28d1812f7163fccdcefc787fda423476e357daec74b2", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ee701ffb4942c6a80acf3e5785df5229bbc3c5a5603c3298c1920f761fcdcbaa", "generation_seconds_batch": 15.69916506903246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.39665150642395, "target_z": 0.5742187269422566, "ce": 5.927301406860352, "genuineness": 4.425758361816406, "identity": 0.5980653762817383, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f06e3c4906bcd4f0925c28d1812f7163fccdcefc787fda423476e357daec74b2.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "f0b053c5902340929c572c1bb800112ae7f70a8dd5074e84ca949ef8770d4cf4", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b44c5b1bdf45aace12f26af6214637557f783e0c594b4a1e88151f27c36a1ad2", "generation_seconds_batch": 15.694387238938361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4065914154052734, "target_z": 0.5820311998585581, "ce": 6.316492557525635, "genuineness": 4.118232727050781, "identity": 0.2690959572792053, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f0b053c5902340929c572c1bb800112ae7f70a8dd5074e84ca949ef8770d4cf4.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "f0ddc4206ecd43bb95a804f7de65a2701b5e208db4c1aca939a15e0bad58be6c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "0924181feef387b4f0df6d51f2bfb8ad01da598cb4fbd612d68e2cb63db2c27f", "generation_seconds_batch": 22.322276729042642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087477684020996, "target_z": 1.1171875622193075, "ce": 6.260893821716309, "genuineness": 4.15588903427124, "identity": 0.5327731966972351, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f0ddc4206ecd43bb95a804f7de65a2701b5e208db4c1aca939a15e0bad58be6c.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f16626ad4d671fb57f606078b1ac35178c8373170c67927d3133dae5ca3bec38", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f510212e4128b03e5bbcc80a91ed3eafdee2e1f7f06c86ae8c16c2c5eaa5a029", "generation_seconds_batch": 16.86037957202643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8511972427368164, "target_z": 0.14550789365959665, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.8144040703773499, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f16626ad4d671fb57f606078b1ac35178c8373170c67927d3133dae5ca3bec38.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "f18d01d828d9e841db9413c7540cdc221a98b7180e22d5620a7b8e73476ce9d0", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "a1b1ecde055c8b4f9c640e5e2b570a49f8d9b5df6b216537eff7664f7a73b69d", "generation_seconds_batch": 23.934990959940478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8772895336151123, "target_z": 0.16601565848862726, "ce": 6.0146708488464355, "genuineness": 4.18099308013916, "identity": 0.6373777985572815, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f18d01d828d9e841db9413c7540cdc221a98b7180e22d5620a7b8e73476ce9d0.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "f1a90a965c03ed6712d613469bcdb4ac7ce257224882304380767e0d81206f3f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "1f6c5c8c0ed6a7567fdf8182628c746ef53b24c037b6354fa8a5211f2bfcfe1b", "generation_seconds_batch": 17.101057136896998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8787388801574707, "target_z": 0.9531250688072341, "ce": 5.855817794799805, "genuineness": 4.507347106933594, "identity": 0.7620620131492615, "hypothesis": "I trouved the petite boîte.", "wer": 0.8947368421052632, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f1a90a965c03ed6712d613469bcdb4ac7ce257224882304380767e0d81206f3f.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "f1d6fdff08f94412fb0f080cda60cac2ed2ae3fd89e3e059b9baf5466b0ea444", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c888013482cd4830d532a2e3e290711f05b404a5ef09118566193845c1093374", "generation_seconds_batch": 26.787865124060772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.3641486167907715, "genuineness": 4.15588903427124, "identity": 0.6212177276611328, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f1d6fdff08f94412fb0f080cda60cac2ed2ae3fd89e3e059b9baf5466b0ea444.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "f1fc627eb4a7b109aca62d79898bda77b2ab070064688924e00546ed34f160fb", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d168672de99aeeeb44cb231a71ef74882695cb08e112582075e63ab15b48c847", "generation_seconds_batch": 15.195250277989544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387514114379883, "target_z": 0.05712881360728115, "ce": 6.117925643920898, "genuineness": 4.068024158477783, "identity": 0.6855147480964661, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f1fc627eb4a7b109aca62d79898bda77b2ab070064688924e00546ed34f160fb.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "7f0f5df555b3957a44a668af1312cc44cf3b3e460eaa597ce849e4865b9bd856", "generation_seconds_batch": 15.167469165055081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.46126127243042, "target_z": 0.6250000819830875, "ce": 6.078212261199951, "genuineness": 4.557555198669434, "identity": 0.7613109350204468, "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_S_MONO_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "f24a00d17f7542450afb747e0e91cbd7ea5556660a7f8cb3db2e2110e3e3e266", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "78e68463344a68074cf3145631a0652c89d0df7fb45a822da40f6f57c755a724", "generation_seconds_batch": 26.512967976974323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.0146708488464355, "genuineness": 4.532451152801514, "identity": 0.672640323638916, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f24a00d17f7542450afb747e0e91cbd7ea5556660a7f8cb3db2e2110e3e3e266.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f2b8cb13049dba3ec596d1f5f370333cc6bb2f10ac05e00389c285a09a97b49d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4099436d75e90ff86fce37fc608dfc8a88d40eb16518654e7af28e62178239f4", "generation_seconds_batch": 21.105477695004083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4850754737854, "target_z": 1.4296876032108512, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.6294920444488525, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f2b8cb13049dba3ec596d1f5f370333cc6bb2f10ac05e00389c285a09a97b49d.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f2e43e59da163a86440b8ba6005e765808cd457369501143cf608dfc0ce365ba", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "e7b62eaaee3030776069cef1faa0da453b82657b53d1d25b421f8b17a3400481", "generation_seconds_batch": 17.693502653972246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.308549880981445, "genuineness": 4.545003414154053, "identity": 0.7004106044769287, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f2e43e59da163a86440b8ba6005e765808cd457369501143cf608dfc0ce365ba.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "f31bf3a531a2402e928e04aa3403415a8ec990ae6b61dafdb126316dcab115d1", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4d81ced4bca80a56e247732d23efaaff17a516dbd42649579ece1802c72c990e", "generation_seconds_batch": 25.246125899022445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4016215801239014, "target_z": 0.5781250570953645, "ce": 6.515059471130371, "genuineness": 4.607763767242432, "identity": 0.3198626637458801, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f31bf3a531a2402e928e04aa3403415a8ec990ae6b61dafdb126316dcab115d1.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "f31c5ae4ed06435687589211cb792466c395369cd7e6a3fb6d67f031d3b94a9e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "6b99969478e105ce87298a1bf53b1cdb07af7072cca169dc2ec36dc0d0c0d11e", "generation_seconds_batch": 17.035671227960847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.027837753295898, "target_z": 1.0703123499416702, "ce": 6.086154937744141, "genuineness": 4.570107460021973, "identity": 0.8941249847412109, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f31c5ae4ed06435687589211cb792466c395369cd7e6a3fb6d67f031d3b94a9e.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "f31c99b445af146e2df95895b75f702124ba80c58e0b7c1a1beb7c2fd1c585b9", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae9a0f306f2650b9bcaa9531430678f5b898bc795271c0ce5a3d1d859372694b", "generation_seconds_batch": 27.780892102979124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.395919322967529, "genuineness": 4.3504462242126465, "identity": 0.4190681278705597, "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_S_MONO_reg/f31c99b445af146e2df95895b75f702124ba80c58e0b7c1a1beb7c2fd1c585b9.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "f377b19ca4530cb1b60dc47c5d780e70912a16277ea2236b731dd6d71bf4667e", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "ae31daf63eeb7831f9263f89697b6da3abd6f2479e21a9aff4c46d445bb2aaaa", "generation_seconds_batch": 24.221133627113886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.037777900695801, "target_z": 1.0781250102478859, "ce": 5.990842819213867, "genuineness": 4.532451152801514, "identity": 0.6529316306114197, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f377b19ca4530cb1b60dc47c5d780e70912a16277ea2236b731dd6d71bf4667e.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "f3ae6b1cc9e2eb514edbad129c7f6b52630ad68b559a62a0171706359b75210a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "52ffbc02bbf8f1043268f2730ab6f55fba8aec0e32140afe5614a6ff2eaeaebc", "generation_seconds_batch": 26.754768899059854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.26743221282959, "target_z": 0.4726562042505092, "ce": 6.316492557525635, "genuineness": 4.319066047668457, "identity": 0.42443224787712097, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f3ae6b1cc9e2eb514edbad129c7f6b52630ad68b559a62a0171706359b75210a.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f462693eb0e26a6c88966e16fd8c01b64fe96bd8d6b55fb49fd8a44f3f11c484", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "55919e00dc403cdc70b7ccd17148c0cccfe74bde90c89b24966d69499a574f75", "generation_seconds_batch": 23.92005821201019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.410112142562866, "target_z": -0.20117188030694094, "ce": 6.181467056274414, "genuineness": 4.532451152801514, "identity": 0.6479122638702393, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f462693eb0e26a6c88966e16fd8c01b64fe96bd8d6b55fb49fd8a44f3f11c484.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "f4a830f85fdf704d7ef82a29a886b3a3fb7ddc4cc871da52a7bd4a13a4fcd7b2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e2ae122e419cad3696e2a6f786c815d46bd33090579e8ed8d746b74d3ab9a211", "generation_seconds_batch": 22.12607882800512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3359761238098145, "target_z": 1.3124999472965866, "ce": 5.6334228515625, "genuineness": 4.319066047668457, "identity": 0.6141819953918457, "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_S_MONO_reg/f4a830f85fdf704d7ef82a29a886b3a3fb7ddc4cc871da52a7bd4a13a4fcd7b2.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f4aaee9c1aa9756e9aea5645ba58a915109928720b9651bc9322340570d04fe8", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "d1bb5a16e107d48cd145685f915314648e3e63c0fea8ece46ee78803064582ee", "generation_seconds_batch": 18.076562891015783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.550720691680908, "target_z": 0.6953125256197148, "ce": 6.125868320465088, "genuineness": 4.162164688110352, "identity": 0.6033000946044922, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f4aaee9c1aa9756e9aea5645ba58a915109928720b9651bc9322340570d04fe8.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "f5807a3bd79a2081ccfccdf6a290562bc9f3bf1ff7c9cd3d2826591cd8514f8e", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "20ac35460d35615d025e69015bfb374c8d433c5bdef9f01f768b14871c3e5b02", "generation_seconds_batch": 18.687576682073995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8698344230651855, "target_z": 0.16015616325896545, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.5595600008964539, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f5807a3bd79a2081ccfccdf6a290562bc9f3bf1ff7c9cd3d2826591cd8514f8e.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "f59545d4ff56a8d5a59b964e4cd5fba25ea2dc0217f718fc754cf60a7cddbc3b", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f49fe70f7a381035e88394fb3365e4d29c7271aa8e76b60ca7f9babb9ff01b73", "generation_seconds_batch": 16.96097723499406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.157057285308838, "target_z": 1.1718750600233319, "ce": 5.982900142669678, "genuineness": 4.607763767242432, "identity": 0.5332908630371094, "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_S_MONO_reg/f59545d4ff56a8d5a59b964e4cd5fba25ea2dc0217f718fc754cf60a7cddbc3b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "f605b306f5c1d7762bfd1dae67678be23cec612de2771f796e01884ff8a795ce", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "346c9d7436a0e66a2effe4b2831d01f01de346f86e1ec33580c752f4dbcc312c", "generation_seconds_batch": 22.633633731980808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.1894097328186035, "genuineness": 4.582659721374512, "identity": 0.6998635530471802, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f605b306f5c1d7762bfd1dae67678be23cec612de2771f796e01884ff8a795ce.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "f61c89ff975de4bbf0345aab4931a5eeafab0d09b86ee1ec50b792777aaeb7e5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d2e1477acfbf99e2f51792454d1f864db5d3a2b11064375f87cde39cd0b7e1c7", "generation_seconds_batch": 27.51203015097417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.655090093612671, "target_z": 0.7773437723257515, "ce": 6.316492557525635, "genuineness": 4.369274139404297, "identity": 0.6193966865539551, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f61c89ff975de4bbf0345aab4931a5eeafab0d09b86ee1ec50b792777aaeb7e5.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "f63eaf25c40ad6b5da94021fcf4610d270326c873cc89c61b0258c6b9ddc470b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "f98fbbd150fc082ce52db60bf1d69f73edcbeea69ebbe79c5507aa1f5ebbb09a", "generation_seconds_batch": 17.376099015935324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.540780782699585, "target_z": 0.6875000527034134, "ce": 6.356205940246582, "genuineness": 4.21237325668335, "identity": 0.29688602685928345, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f63eaf25c40ad6b5da94021fcf4610d270326c873cc89c61b0258c6b9ddc470b.mp3" }, { "id": "vn_S_MONO_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, 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_S_MONO_reg-de-0", "key": "f66c17f99e9fc271c3fc7b16f86c019dae341bd9a9d49530ab27f017dd74195d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "babe90529d37498decfe0ccd02c64d530baa28b31d9a4e17ef606529152b5e02", "generation_seconds_batch": 23.57065650902223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6501200199127197, "target_z": 0.7734374421726437, "ce": 6.006728172302246, "genuineness": 4.457139015197754, "identity": 0.6455991268157959, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f66c17f99e9fc271c3fc7b16f86c019dae341bd9a9d49530ab27f017dd74195d.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "f693fa447478e1751bb63eaeebbc7c3715c03f9225ee09adfb25d69e545d1e4b", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "4a09be323e419d0ccb2e16c52ace682a79a6f2425a8d847096b8a9d98ea3fab7", "generation_seconds_batch": 16.938465461018495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9617791175842285, "target_z": 0.23242195936206103, "ce": 6.062326908111572, "genuineness": 4.425758361816406, "identity": 0.5252628326416016, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f693fa447478e1751bb63eaeebbc7c3715c03f9225ee09adfb25d69e545d1e4b.mp3" }, { "id": "vn_S_MONO_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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_S_MONO_reg-nl-0", "key": "f6a2539d963ca462054295871940fabc8800dec323bf2838cc6d0d7e3f901140", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c2845e025ee4a75a2a300f5b64993cdcfe7b853d9634038173a9b3ba9b69a524", "generation_seconds_batch": 22.671564646996558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6878092288970947, "target_z": 0.017089772792539783, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.6023052930831909, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f6a2539d963ca462054295871940fabc8800dec323bf2838cc6d0d7e3f901140.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "f70d9fc2b03eee857cd83164266a478bfd1ab4b881c8396632f5ca51260b66a5", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "c03625c8c6a6c1c6c821a8bf1f41f4128a09fbbf4759a55d07b5afd45415103e", "generation_seconds_batch": 22.322276729042642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.928438663482666, "target_z": 0.9921876207786557, "ce": 6.23706579208374, "genuineness": 4.21237325668335, "identity": 0.32589006423950195, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f70d9fc2b03eee857cd83164266a478bfd1ab4b881c8396632f5ca51260b66a5.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09d0005afd85701dda716d1a19a5d0e56f62070a5252f656b60c752e71693157", "generation_seconds_batch": 19.658429977018386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6053903102874756, "target_z": 0.73828122035433, "ce": 5.95112943649292, "genuineness": 4.218649387359619, "identity": 0.8273202180862427, "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_S_MONO_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "f80a90f8963bad40d69194eb73a7aedd103ab2b744f996459b09121c9aa11734", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "108a0dc48d665440a9f23dca94c5d6bfd0eee8a8b3156feb9edaa91ef1627c37", "generation_seconds_batch": 16.003276851028204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8787388801574707, "target_z": 0.9531250688072341, "ce": 6.284721851348877, "genuineness": 4.243753433227539, "identity": 0.5366246700286865, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f80a90f8963bad40d69194eb73a7aedd103ab2b744f996459b09121c9aa11734.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "f8329cfa0122ac65f950223f1d749e578c39f4b4f7c28c090605eb047c93a63a", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b5d5b90f48c780e1d70486c33f494568cc5ae70f7d6515708cdadee68e6c0fc0", "generation_seconds_batch": 22.74904941197019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.186877250671387, "target_z": 1.1953126661621505, "ce": 6.062326908111572, "genuineness": 4.093128204345703, "identity": 0.21690510213375092, "hypothesis": "Ik 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_S_MONO_reg/f8329cfa0122ac65f950223f1d749e578c39f4b4f7c28c090605eb047c93a63a.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "f8542d66b382ac1d07ff9e4daa51b00504270db833f3ddbced46c3f1bc65c370", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e1d5a4cb635ca217c8083366011c81a061e4161f8037af514ad9556001a8be32", "generation_seconds_batch": 22.92863033595495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.634174346923828, "target_z": 1.5468748843452873, "ce": 5.967014789581299, "genuineness": 4.256305694580078, "identity": 0.7899383902549744, "hypothesis": "La porte 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_S_MONO_reg/f8542d66b382ac1d07ff9e4daa51b00504270db833f3ddbced46c3f1bc65c370.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_S_MONO_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": "41baf421912fdf73a6211d777bb2a57f2ef0f5b17aa30412d4a870e13f3ba53e", "generation_seconds_batch": 19.124837178969756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5159308910369873, "target_z": 0.6679687767177026, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.5664417743682861, "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_S_MONO_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "f8853b17ee0617253fb1d85bf70692bc3cfc80e54b61025a75a068c0fc399862", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0c6fe99bdaff21e804de482541a5ff69814f06ae9e0c6c8b9d8031493089e194", "generation_seconds_batch": 17.024466399103403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.421501398086548, "target_z": 0.5937500029279674, "ce": 6.205295085906982, "genuineness": 4.3504462242126465, "identity": 0.7782323360443115, "hypothesis": "La porte 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_S_MONO_reg/f8853b17ee0617253fb1d85bf70692bc3cfc80e54b61025a75a068c0fc399862.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "f89dbbeb3795fd9239c600643a6424691a64e82f4de8760d7704e84a7fcc9d4e", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9451872cb4e369b34e36230c2978389cb020e98bb7cb2281f218ae4a2e29731c", "generation_seconds_batch": 27.10434698499739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8389790058135986, "target_z": 0.921874989752114, "ce": 5.9431867599487305, "genuineness": 4.595211505889893, "identity": 0.8033635020256042, "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_S_MONO_reg/f89dbbeb3795fd9239c600643a6424691a64e82f4de8760d7704e84a7fcc9d4e.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "f8b962bc340a92060e7757e939a7fede64c3ca54647ee77e2dfb2a622058efb7", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5979cbd9c0064ccab29831b2bf32b7dbf29d58586aeac1f8fc9a59b395bb719", "generation_seconds_batch": 20.4519708189182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.157639026641846, "genuineness": 4.6705241203308105, "identity": 0.7149758338928223, "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_S_MONO_reg/f8b962bc340a92060e7757e939a7fede64c3ca54647ee77e2dfb2a622058efb7.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "f8f7fda7a13ce35d74f0722e8da6955a60a90c49c27a1526a700be7c9cd6cbca", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "80b185e551f8ce38f623c7b7bf99d60fdb5a91c24f084517fba8464a815cb581", "generation_seconds_batch": 17.61940596299246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.419747352600098, "genuineness": 4.074300289154053, "identity": 0.4811300039291382, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f8f7fda7a13ce35d74f0722e8da6955a60a90c49c27a1526a700be7c9cd6cbca.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "f95e743d533c64c29815392931dedc545f995506dc34c38032ddb53d452ee19c", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "47816f2ec015d9de323a92886972ab91ff2b0471ce8356940801c3a03f78f5f5", "generation_seconds_batch": 18.102993670967408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.348263263702393, "genuineness": 4.413206577301025, "identity": 0.3715018332004547, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/f95e743d533c64c29815392931dedc545f995506dc34c38032ddb53d452ee19c.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "fa0b1dc7e6c8ca8f0d744c130709e8507bba4e75285912166e0de2f08f303704", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "37e42ce9a891c2c3c654b0e066ed3bc431571410e85cc0c5e7116d62aed23959", "generation_seconds_batch": 20.742347405059263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7495195865631104, "target_z": 0.8515625461154868, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.6917271614074707, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fa0b1dc7e6c8ca8f0d744c130709e8507bba4e75285912166e0de2f08f303704.mp3" }, { "id": "vn_S_MONO_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. 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_S_MONO_reg-de-1", "key": "fa4ffc3b8ca637d0ceeeabf85d1b43d1f780638033c530f483f4cfb08bbf93da", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "07eee3778aa9a9811b5a7c73bb30e551d09a665f7f77d1bada433cdf07a09bca", "generation_seconds_batch": 16.60791674396023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.435632705688477, "genuineness": 4.293961524963379, "identity": 0.5995075702667236, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fa4ffc3b8ca637d0ceeeabf85d1b43d1f780638033c530f483f4cfb08bbf93da.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "fada11a0d10375b08dea511fae972f685d2bdccc391e0f5c7030278e01cd86d9", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "85bdb39144294491e7630c14c534486e1718b6d8f484218e0966731c79882208", "generation_seconds_batch": 16.237013614038005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8787388801574707, "target_z": 0.9531250688072341, "ce": 6.062326908111572, "genuineness": 4.143336772918701, "identity": 0.5875276923179626, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fada11a0d10375b08dea511fae972f685d2bdccc391e0f5c7030278e01cd86d9.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "fada2abce819d7212b9246792f87edb190f1837e600e68a751cf5e4c49fbb7e7", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "91ccfe07e6c95f37987e6479a9c18e59e08ce605861356cff14e9c41b8986cfc", "generation_seconds_batch": 18.134099129936658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0412986278533936, "target_z": 0.29492193008238693, "ce": 6.300607204437256, "genuineness": 4.469690799713135, "identity": 0.3205278515815735, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fada2abce819d7212b9246792f87edb190f1837e600e68a751cf5e4c49fbb7e7.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "fb188f9f1d71c5f62c068c677d4cdeb2e1ec0195f0831565518edb2f707ddef3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "56a7414197e846117645153103f62acf629932c63b839d067bf04c95ce33f5fd", "generation_seconds_batch": 29.258746451931074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.723633766174316, "target_z": 1.6171873279819147, "ce": 5.621508598327637, "genuineness": 4.494794845581055, "identity": 0.8844459056854248, "hypothesis": "La porte 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_S_MONO_reg/fb188f9f1d71c5f62c068c677d4cdeb2e1ec0195f0831565518edb2f707ddef3.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "fb27e882b987c55b872adc5b4bb9abba7fbdddbe80eb0f5745ffd0448fc7de45", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "546c4e8a69774c8ec1cb0c7473f7f257b802029340fca0a61339cc5a40cc4a62", "generation_seconds_batch": 27.68967806000728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614294528961182, "target_z": 1.5312499385126843, "ce": 6.070269584655762, "genuineness": 4.482243061065674, "identity": 0.7921208143234253, "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_S_MONO_reg/fb27e882b987c55b872adc5b4bb9abba7fbdddbe80eb0f5745ffd0448fc7de45.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "fbe18424b6d8c55158f82d0f37f9f83c5a36556f5c4eefc1db2e31b7f6f8e247", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8ad964767b92557d942fe616379893ea368139a6e91191bdea652e8e04b6a9f", "generation_seconds_batch": 19.921126255998388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.893648624420166, "target_z": 0.9648436844867292, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.5865263938903809, "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_S_MONO_reg/fbe18424b6d8c55158f82d0f37f9f83c5a36556f5c4eefc1db2e31b7f6f8e247.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "fbfa6663502057b07374f497759955809ac11aec6185468c473e5f0b2700648d", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "967999aa5a90747c52e2832ee373e458ea2fe2d4cc46d794ff8d1065a5648dc2", "generation_seconds_batch": 30.363590634893626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3270719051361084, "target_z": 0.5195312291382322, "ce": 6.141753673553467, "genuineness": 4.080576419830322, "identity": 0.5714753866195679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fbfa6663502057b07374f497759955809ac11aec6185468c473e5f0b2700648d.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "fc1214279250f7ff37a32c76b497b223637b74ba495f239e0e4a696a7b30784c", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b0f920396ea01d9c0bcd02ac1e03f3e43f4975c121f6fb7a1a9fa343711168ae", "generation_seconds_batch": 22.60140514408704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007957935333252, "target_z": 1.0546874041090673, "ce": 6.197352409362793, "genuineness": 4.3127899169921875, "identity": 0.5742720365524292, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fc1214279250f7ff37a32c76b497b223637b74ba495f239e0e4a696a7b30784c.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "fc57e5519d5b34670558c134dcc6e33ab370e6dbd8fa68c9e6a1f1542b173714", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "631bdc4b9addca0591d2eec97f9aa18a814e9eff6fb60a2cbbae7c3be3adb4e2", "generation_seconds_batch": 24.81174633000046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.615330219268799, "target_z": 0.7460936932706315, "ce": 6.117925643920898, "genuineness": 4.105680465698242, "identity": 0.49450480937957764, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fc57e5519d5b34670558c134dcc6e33ab370e6dbd8fa68c9e6a1f1542b173714.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "fccf41cc078d56dfa1735e881d071b9cd8ed42f94165b0aca8bfcdd7ff0c34bc", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "b2afc36b9a059469d877497bf8b8ce03e396ac5639bf9134c7f2398ae8b03fb5", "generation_seconds_batch": 24.34437019797042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.166996955871582, "target_z": 1.179687345549719, "ce": 6.125868320465088, "genuineness": 4.143336772918701, "identity": 0.6169878244400024, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fccf41cc078d56dfa1735e881d071b9cd8ed42f94165b0aca8bfcdd7ff0c34bc.mp3" }, { "id": "vn_S_MONO_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_S_MONO_reg-nl-1", "key": "fcf0c16fb492d4e89aec3aed9989d3585970fccb90675947ee00bef1855ab194", "bucket": "vn_S_MONO_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "294001005a95a66e4d1dbdd708b5ef16eda030d6cb4b18eab8e9c47c41a0ad2c", "generation_seconds_batch": 18.076562891015783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1829428672790527, "target_z": 0.40625009076698976, "ce": 6.181467056274414, "genuineness": 4.105680465698242, "identity": 0.5456105470657349, "hypothesis": "Ik 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_S_MONO_reg/fcf0c16fb492d4e89aec3aed9989d3585970fccb90675947ee00bef1855ab194.mp3" }, { "id": "vn_S_MONO_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We 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_S_MONO_reg-en-1", "key": "fcf91374a03e23d8465a815a4ca76f33a62e6eae1861ece5c4682333a85e412f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "44f26bd76433dfd8b3d1f4c506eb25db2d273011565337152e900f0d9a8d483f", "generation_seconds_batch": 24.199242640985176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5556907653808594, "target_z": 0.6992188557728227, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.36689493060112, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fcf91374a03e23d8465a815a4ca76f33a62e6eae1861ece5c4682333a85e412f.mp3" }, { "id": "vn_S_MONO_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me 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_S_MONO_reg-en-0", "key": "fde2a707f155091d9e222d5b7afd4ae350a7b7df7438b5451d20420f0dfa265b", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "24183a9ec4da5b944faee8449d321df3c4fd3cb48ab103d05b98004666aaebaf", "generation_seconds_batch": 25.76906446600333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.3245808184146881, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fde2a707f155091d9e222d5b7afd4ae350a7b7df7438b5451d20420f0dfa265b.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "fde9c17c4b519fffeb9dfdfc6022958a7a378a5df0921b60e1866253a2f9593d", "bucket": "vn_S_MONO_reg", "phase": "heldout", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c3547cf93ab7b8c369782c5d51f41246404a84cdd25fe110fa0fa85cbc0fbacc", "generation_seconds_batch": 27.45250523800496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858859062194824, "target_z": 0.9375001229746313, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.5288660526275635, "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_S_MONO_reg/fde9c17c4b519fffeb9dfdfc6022958a7a378a5df0921b60e1866253a2f9593d.mp3" }, { "id": "vn_S_MONO_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. 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_S_MONO_reg-es-0", "key": "fe07c672120388124fdf41f9a393147ad3dc083a23c1520a6affd5770ff77c15", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "67f58f35f245929dd071dc8aef2e71884ad3dacb745b4385d1086b30c5532599", "generation_seconds_batch": 22.828428993932903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5258708000183105, "target_z": 0.6757812496340041, "ce": 5.95112943649292, "genuineness": 4.231201171875, "identity": 0.6210335493087769, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fe07c672120388124fdf41f9a393147ad3dc083a23c1520a6affd5770ff77c15.mp3" }, { "id": "vn_S_MONO_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi 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_S_MONO_reg-fr-0", "key": "fe52fd37ed415fa1f59efbb45b20b48b1d9fc9f6cbdca2ccbb30d0f46164e5f2", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1cf09721f413e6b13d5658d6a775aa2d6895ca32c08f7c683c5548917121427f", "generation_seconds_batch": 21.765763310017064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.431441307067871, "target_z": 0.6015624758442688, "ce": 6.2767791748046875, "genuineness": 4.218649387359619, "identity": 0.7066093683242798, "hypothesis": "La porte 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_S_MONO_reg/fe52fd37ed415fa1f59efbb45b20b48b1d9fc9f6cbdca2ccbb30d0f46164e5f2.mp3" }, { "id": "vn_S_MONO_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble 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_S_MONO_reg-fr-1", "key": "fe5a2e98583aff941c4cfc8f18c32e97aeed6c8f071534acb4a283e5872beac3", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "cae0107526cb625f4eac4e6f06246cee5ddd5faecb201ef845f6ff63ed2eca58", "generation_seconds_batch": 25.13087231805548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.693814277648926, "target_z": 1.5937500966229246, "ce": 5.959072113037109, "genuineness": 4.7081804275512695, "identity": 0.7806000113487244, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble.", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/fe5a2e98583aff941c4cfc8f18c32e97aeed6c8f071534acb4a283e5872beac3.mp3" }, { "id": "vn_S_MONO_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos 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_S_MONO_reg-es-1", "key": "ffd46c32873e420b51659bda3d950df0bacea45aebabe5f9f36d3688bcd7eb5f", "bucket": "vn_S_MONO_reg", "phase": "grid", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(An inward, connected spoken monologue, thinking aloud with natural expressive pitch and one continuous personal train of thought. 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": "17a7c986df1a4d70cb7b30195cd8e2152e48d77b2416cea614773e3297f46724", "generation_seconds_batch": 21.278126031975262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4861111640930176, "target_z": 0.6445313579687982, "ce": 6.221180438995361, "genuineness": 4.68307638168335, "identity": 0.5632830262184143, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_MONO_reg/ffd46c32873e420b51659bda3d950df0bacea45aebabe5f9f36d3688bcd7eb5f.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "0f4c12d3e9cdf260655ec9e4f42a6466a8ff0da69ae6010828c1e2fa503c6faa", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.232642412185669, "target_z": 0.445312455348497, "ce": 6.109982967376709, "genuineness": 4.457139015197754, "identity": 0.6560628414154053, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/292152688d79ad06ea699dea.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "ce434822cbdc029b27f6f7e17c6db300bef908992f5554d8a8f1b58638232772", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.153122901916504, "target_z": 0.3828124846281711, "ce": 6.2767791748046875, "genuineness": 4.256305694580078, "identity": 0.5665228962898254, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/22a210e789d0d60403d051f8.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "a726b0c3f9550dfe615a287af4bc99c39a23adc458e91fc14317fd896be829cd", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.0462684631347656, "target_z": 0.2988280728455805, "ce": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.4675672948360443, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/2ef512a26451d956d6872ccc.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "efe8f94c35f7b0580a2194b89d0ff103b2bb24cc4eb0b6f7dd96788e8bc486a0", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.3320419788360596, "target_z": 0.5234375592913401, "ce": 6.205295085906982, "genuineness": 4.137060642242432, "identity": 0.392148494720459, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/253bb86613677cc627580dd1.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "00ff2aaf133cd1db7a6e2d28a0db102d3d88b12b973449bc3a083b94719ef975", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2400975227355957, "target_z": 0.4511719505781588, "ce": 6.284721851348877, "genuineness": 4.582659721374512, "identity": 0.7704691886901855, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/6703931e3c4c310e6e51b066.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "9fd301606d1b46b007bee9777a027a1a7784dbb0196f549b63c9eb9056b209f9", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.28482723236084, "target_z": 0.48632817239647247, "ce": 6.340320587158203, "genuineness": 4.3504462242126465, "identity": 0.4932088255882263, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/c8da5095dabed1ef8690b55f.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "4a330f5cc7fb72407d53cf7c0636f4004d13796a446335fdab2763b8c8efcd76", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.038813591003418, "target_z": 0.29296876500583297, "ce": 6.229123115539551, "genuineness": 4.331617832183838, "identity": 0.4661012887954712, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/eedb1f9612058f8989354860.mp3" }, { "key": "original-vn_S_MONO_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "a537b89ce440283c9e0dbf811c2e464c8481dde73ce06dba28a7111b9b0cdf4b", "task_id": "vn_S_MONO_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.277372360229492, "target_z": 0.48046886455672494, "ce": 6.125868320465088, "genuineness": 4.231201171875, "identity": 0.3927881717681885, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/3690c9b881eb8546dac9dce3.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "e3c02ab9aed48a475181b37c1e9dafadc0e427850d613b30de420f0523af1511", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6053903102874756, "target_z": 0.73828122035433, "ce": 6.395919322967529, "genuineness": 4.256305694580078, "identity": 0.57170569896698, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/199aaf800ef7d418f45161e6.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2eb7dc13e56eb3b3208efb44dfabdab839711a7e3a59944065b93bda5258f2b3", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.443574905395508, "genuineness": 4.049196243286133, "identity": 0.6008443832397461, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/d4804d45469f2934360d447c.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "04b56bef6ed816bb0b8c7a9c0b80ab8e777b4e71b339b525d2a64623d31e0e40", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7594594955444336, "target_z": 0.8593750190317881, "ce": 6.4515180587768555, "genuineness": 4.080576419830322, "identity": 0.5134449005126953, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/49a2cff2517ce672648451fb.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "f04544e9be88f19a53e2e1d4658abf32be3f5c5da19968bd2dbfafd2340d7d92", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.4515180587768555, "genuineness": 4.494794845581055, "identity": 0.537783145904541, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/42f18dc237cff0bc11839516.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "233bfdd2fe4f39b393cc2cd2ed1d5b1f852bc47fe8450dd8b401717c4bbca935", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9085586071014404, "target_z": 0.9765624875561385, "ce": 6.419747352600098, "genuineness": 4.20609712600708, "identity": 0.32241255044937134, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/017e9058fcc8e46056f6fcfe.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "467ad0beb629f306b494ce75b0cad3bcf51a37fcada8650e85e944929d4bf7b5", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.928438663482666, "target_z": 0.9921876207786557, "ce": 6.348263263702393, "genuineness": 4.432034492492676, "identity": 0.5860497951507568, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/0f0cfc921b0fb3199209ea9d.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "81a38481d1c4c4f934ec4e8c7928b8516ce8ab5c50072007f66b9c50fe799e99", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 6.395919322967529, "genuineness": 4.2751336097717285, "identity": 0.5980714559555054, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/d5204f1e6863c7cc24180cf2.mp3" }, { "key": "original-vn_S_MONO_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "5bb1a82994cbc453b3a150427e81c93c70456d34b6ec4dea342a9d4105ebe588", "task_id": "vn_S_MONO_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660059928894043, "target_z": 0.7812499150889451, "ce": 6.332377910614014, "genuineness": 4.281409740447998, "identity": 0.6620054244995117, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/de/4e3680a4ff17c2a7b18d574f.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2fb1d9807fa5de392b4a26a826da66f8a4b379ea652d2dc11aadda84aa9c4f04", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.346951961517334, "target_z": 0.5351563623607494, "ce": 6.181467056274414, "genuineness": 3.5345609188079834, "identity": 0.29820048809051514, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/ccc986d89edee9083b298bbd.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "e5bfe10d888348ad5360ca58f08ddbb528dfd072212c36b45f87b58f8a5ae7fe", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.351921796798706, "target_z": 0.5390625051239429, "ce": 6.38797664642334, "genuineness": 4.068024158477783, "identity": 0.4063401222229004, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/f4b6a762bb11dc8d60138cdd.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "cc5b8f08479801265c4ae9a7f0646eb2b0bb9a0a7c640d922aba313ac52ec389", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5606606006622314, "target_z": 0.7031249985360163, "ce": 6.125868320465088, "genuineness": 3.6412534713745117, "identity": 0.37847936153411865, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/403945636d1b3e8d88df7624.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "b842bd64c78ce320890b00db55d9eb02dded3b178beb487def0b75dc68d6ed96", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.0412986278533936, "target_z": 0.29492193008238693, "ce": 6.300607204437256, "genuineness": 4.017816066741943, "identity": 0.27583637833595276, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/315ca2a9d804d186da9acf23.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "153d1274ab91a3b7a62de2981b9fc0c85ed0df429f04fd1b63ebc33ee1372f84", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.7468278408050537, "target_z": 0.06347664695356, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.34686851501464844, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/ffff73dd435cc2ec4bdaf5bc.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "57e833ec8ba2ca9dbda6f128a941a7ea2748530a7c726abcc1eda1c7f8334226", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.4065914154052734, "target_z": 0.5820311998585581, "ce": 6.125868320465088, "genuineness": 3.9676074981689453, "identity": 0.17420151829719543, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/ef153289c5f80d8a800ffeab.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "9c7d697a32d7e095fdd9269f4d93d57fb45b38ab564d248f6f04c0323a0a63e4", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.351921796798706, "target_z": 0.5390625051239429, "ce": 6.324435234069824, "genuineness": 4.256305694580078, "identity": 0.18615445494651794, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/046f62b7090bbd4c33657743.mp3" }, { "key": "original-vn_S_MONO_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "79dc7b9a4e83e3db3db37eed8a5313dbd4a9952d0fea2a652aca11f4620d16dd", "task_id": "vn_S_MONO_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.9184985160827637, "target_z": 0.98437496047244, "ce": 6.173524379730225, "genuineness": 4.20609712600708, "identity": 0.17589402198791504, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/b777d876c3b4d91bf9ec967e.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "fde69a6114ce665f01dc3bc4d5bd513a6794d1a16fb247bb7277374cbced7ef8", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.8089523315429688, "target_z": 0.11230464952792264, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.5002711415290833, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/a40d8c39d19a5fdaf83ce558.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "fe13ec860194c16238580402769f2ea34037e3c03ec1212b3004b6bceef0d5a1", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.510960817337036, "target_z": 0.6640624465645948, "ce": 6.260893821716309, "genuineness": 4.281409740447998, "identity": 0.492092490196228, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/fdea58a4dbe24e5a4ad23d06.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "8b9166a2dc4236861f08c63513800114d938a5b758744ed60a697669c56afc19", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.395919322967529, "genuineness": 4.293961524963379, "identity": 0.4126124382019043, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/7fba6947bc6fed4b9002a0e1.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "967ab88060b78f5b1ee3e19019a9b4c9a215254f451f1281363fe5d2f734f3a7", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.0189337730407715, "target_z": 0.2773438191732301, "ce": 6.308549880981445, "genuineness": 4.406930446624756, "identity": 0.39197343587875366, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/bfc689048a67608c20fc4584.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "80ba22a026a3085394e18a8ae5ae0fc4cd2cbeaf55220f24623be607fa4a08b5", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.317131996154785, "target_z": 0.5117187562219307, "ce": 6.348263263702393, "genuineness": 4.406930446624756, "identity": 0.44432950019836426, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/3f75d2077929195619786ecf.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "fd0803cde97350a4b3681a682e57f6498d19a42df981609493d85bbdc199cf9f", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5954504013061523, "target_z": 0.7304687474380285, "ce": 6.260893821716309, "genuineness": 4.325342178344727, "identity": 0.3302083909511566, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/c6a1a5b50e9835ba2a0d2355.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "772fb7b88c244a1348c9564e3d46b8e62fef5dd0071ab9cba5edcad9ab704949", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.272402286529541, "target_z": 0.47656253440361707, "ce": 6.372091293334961, "genuineness": 4.3881025314331055, "identity": 0.34778928756713867, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/f742982add093d9a14b2efa7.mp3" }, { "key": "original-vn_S_MONO_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "6834967f84ed6bbf76dcf1401833c1e5ff695b050fdbf55d2b27863fef304e22", "task_id": "vn_S_MONO_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.3618617057800293, "target_z": 0.5468749780402444, "ce": 6.38797664642334, "genuineness": 4.532451152801514, "identity": 0.31341761350631714, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/en/0990a377483e94123cf6f403.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "536ca341f58d51767750b43a61cdebb9b02da90f738573662e2fc9f830e8ea58", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.739579677581787, "target_z": 0.8437500731991853, "ce": 5.998785495758057, "genuineness": 3.7542223930358887, "identity": 0.5846163034439087, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/0e62e88df09d6e73750619d1.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "71a63f6b4ae5b90ad89cb400e596599ae97b0b61bb3f5d53b6499e3e9540f602", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 5.879645347595215, "genuineness": 4.281409740447998, "identity": 0.6607774496078491, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/ed7142245a53207c98566497.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "ef77ed7f30f75a134945a6916e9c25baac6bcd441275fe136828ba7441e61095", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7892794609069824, "target_z": 0.8828126251706068, "ce": 6.038498878479004, "genuineness": 4.231201171875, "identity": 0.709069550037384, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/4acc1d586800d87d40fb6223.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "20f67d85662cd252b54751e631ff4ce392126bb68320b39256479b13c8648fa9", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8737688064575195, "target_z": 0.9492187386541263, "ce": 5.871703147888184, "genuineness": 3.9927117824554443, "identity": 0.623967170715332, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/1c8aebef171e89cc1ca7deed.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "8004960d7b540c892a204090865cfc30a6f6af6b054861e99e4854bf1763352f", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9781382083892822, "target_z": 1.0312499853601629, "ce": 5.967014789581299, "genuineness": 4.356721878051758, "identity": 0.6461224555969238, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/9f34f84cb05e0fd5a5b567be.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "a27ac00d8550b0fa45f7f63365916736d152893b8d80523617cec25921119b42", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 5.871703147888184, "genuineness": 4.124508857727051, "identity": 0.6032805442810059, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/aebd4d23290b2198b2fc3fd3.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "cf7e41156a2d9a94fbea9b91788b7dce310c19a2f69c14c350b5d5e32038260f", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.769399404525757, "target_z": 0.8671874919480896, "ce": 5.855817794799805, "genuineness": 3.9927117824554443, "identity": 0.6778576374053955, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/468261505114727b716814f4.mp3" }, { "key": "original-vn_S_MONO_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "279315e9579a74298465b134c27ee58c0bb2b67f09f1bbc0641b4a73bbf84320", "task_id": "vn_S_MONO_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6053903102874756, "target_z": 0.73828122035433, "ce": 6.046441555023193, "genuineness": 4.250029563903809, "identity": 0.4435858130455017, "hypothesis": "La puerta sigue abierta y he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_MONO_reg/es/031887a76ca9f44e849b68d9.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "30a22ca4da5dd98102ed7ad6472c660dce1c2d76c4075be557cb42ff0e0b2425", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.819099187850952, "target_z": 0.9062500439195111, "ce": 6.165581703186035, "genuineness": 4.111956596374512, "identity": 0.7593462467193604, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/cdd88145a1c778e397b52175.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "953a10e54359d59dbd86583c7dd84e249a89f517cc378920bc01db469f8aa5e3", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.431441307067871, "target_z": 0.6015624758442688, "ce": 6.1894097328186035, "genuineness": 4.796044826507568, "identity": 0.7171478271484375, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/5c1cbbd58bc897138c22ce9e.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "ed5c98337c9f6c83b7ed5c372008882a5a0accea97cbb245a1a457bb2de23c09", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.580540418624878, "target_z": 0.7187499443686192, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.7239075899124146, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/be22b4dba645bdd7a59b8eb1.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "305e68107132ac9c998dfebb16c207d1da54d3566cf6827dd6703ccbb488a37a", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.819099187850952, "target_z": 0.9062500439195111, "ce": 6.0305562019348145, "genuineness": 4.281409740447998, "identity": 0.5359903573989868, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/6ac72d5a94d95801bdade66a.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "d57ebfa47f542e424bdf680b73d6f16fc4584ae3a15e7d87b5cbff27bf9272e5", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.5447241067886353, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/fda654458588401e802b1c78.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "be57207c0c6e8b637f02b043c0354cf7f17ae2f8b9d4b5610e7273bb489870db", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.42647123336792, "target_z": 0.597656145691161, "ce": 5.927301406860352, "genuineness": 4.68307638168335, "identity": 0.7658534049987793, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/46947579582cb7fca826d16c.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "582738ee91957ab9f4ebabeafcc6cb3961c44e34e5b075b5ad04928b4a61b710", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.165581703186035, "genuineness": 4.231201171875, "identity": 0.6270463466644287, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/53eed1438ef97f41b4dba96e.mp3" }, { "key": "original-vn_S_MONO_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "699ec63698b38e998398978dfc5fa84ed81e744d044de2f6be3dac3d9d9563fb", "task_id": "vn_S_MONO_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 4.057657718658447, "target_z": 1.0937499560804889, "ce": 6.324435234069824, "genuineness": 4.632867813110352, "identity": 0.6274929046630859, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/es/fe3176bd300acecaf815d692.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "88ac341323bd0ae1d03023896a61fbc123878d7f3f7fb9f02ff22e73f818c290", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.590480327606201, "target_z": 0.7265624172849207, "ce": 6.181467056274414, "genuineness": 4.331617832183838, "identity": 0.6958494186401367, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/4512be784287c543fec8d314.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "b4c96955d0420fe0ccf2c3b000eb58cbeb78995019db936a837784a94a0d5179", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 4.017898082733154, "target_z": 1.062500064415283, "ce": 6.006728172302246, "genuineness": 3.9801597595214844, "identity": 0.7659323811531067, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/d24f02cca4c1ecd069ca45c0.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "9e030d82cec65045c493363da89e48b8f2f5e815fe6900d930934599bb8e57f9", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.6700000762939453, "target_z": 0.7890625753951608, "ce": 6.221180438995361, "genuineness": 4.243753433227539, "identity": 0.8142993450164795, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/58b945166cde9c1710447357.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2d212b103cb5d844eb1cf67552e0041aa15bede4c5032fcee695b9f414245fbe", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.7992191314697266, "target_z": 0.890624910696994, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.6598905920982361, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/8d85435dfffc36a5f30c2a66.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "c6e01e267ec2d73d3178ddea6f0c8c35459588eeaf6a4157a3f0018db194009a", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.386711597442627, "target_z": 0.5664062540259552, "ce": 6.197352409362793, "genuineness": 4.532451152801514, "identity": 0.590533971786499, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/68ee1178654863d4e4d6aa7b.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "aa57dde9f9433848ce0ac7b92a68886205e916b771c184afc1e6bc629378b826", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5308408737182617, "target_z": 0.6796875797871119, "ce": 6.141753673553467, "genuineness": 4.494794845581055, "identity": 0.6350662708282471, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/cee2d9b114bb2bc11e67a0b8.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "c31f5faa7ffae47e0844b2da12688ce18437077b69452ac6a962fe1733a929f7", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.056208372116089, "target_z": 0.30664054576188193, "ce": 6.173524379730225, "genuineness": 4.645420074462891, "identity": 0.7698993682861328, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/0c2aa43e98d9b552f2e51e08.mp3" }, { "key": "original-vn_S_MONO_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "9ecd43fbbcc39c4e64804a39010318ecf3fab672a5458def37ea94922c414aa3", "task_id": "vn_S_MONO_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.699819803237915, "target_z": 0.8124999941440652, "ce": 6.046441555023193, "genuineness": 3.8922953605651855, "identity": 0.7170294523239136, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/c2e4dffcf97d311d79ee66cb.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "e728ee5968c462b74fec145f599f8243b685c9ab34712a9bb44357cecf7a0f19", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2450673580169678, "target_z": 0.45507809334135235, "ce": 6.078212261199951, "genuineness": 4.645420074462891, "identity": 0.61588454246521, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/2ec6687b34431ca662a3da18.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "55d19ac86b09d7aabde550923e3730a20d00ea411324355e5f15993800269e4d", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5954504013061523, "target_z": 0.7304687474380285, "ce": 6.0305562019348145, "genuineness": 4.344170093536377, "identity": 0.802674412727356, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/66bb5abcd192f6c74a656b8c.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "379d352c90f69b2c13b91c5c2ce48f6ce8e518bc381a8c133d6669cd35e84f0d", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.4513211250305176, "target_z": 0.6171874216768718, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.7773863077163696, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/09e55613821df08d26355430.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2aecdfc81db48812ef3c510c346b6ca3929c3fc3bd4c42d0a9cb42e41036ca24", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.9582583904266357, "target_z": 1.0156250395275601, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.7343493700027466, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/cdc10ebe1f0e33f55edc69bc.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "3b6911022e1bf4284c8a479b78227c652de68cf680cc867ff26ba0f13684dbf6", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.77933931350708, "target_z": 0.874999964864391, "ce": 6.1020402908325195, "genuineness": 4.519899368286133, "identity": 0.7712496519088745, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/6545bb4ee1c242c4f21e6fe3.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "0474354449475eb3e4bb9472def778ab6ac39003108e4ecd5b0db0ca5f8d0860", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.8687987327575684, "target_z": 0.9453124085010184, "ce": 6.141753673553467, "genuineness": 4.061748504638672, "identity": 0.6110432147979736, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/f24f2c788918326414acbe65.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "50fb2d30a24a98c1b237ed45dcd3b79bd7817a9c0226830bbe4eff185f23b705", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5209007263183594, "target_z": 0.6718749194808962, "ce": 6.213237762451172, "genuineness": 4.7332844734191895, "identity": 0.8280959129333496, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/304a251ae4d2f4e7f4490294.mp3" }, { "key": "original-vn_S_MONO_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "128003c0ef88510963c850dba88960271f3fb4553666217bc82dcd932be9c42e", "task_id": "vn_S_MONO_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.6501200199127197, "target_z": 0.7734374421726437, "ce": 6.173524379730225, "genuineness": 4.808597087860107, "identity": 0.74504154920578, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/fr/8d3ca86380ae885a1bca4aef.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "37b5d5c381de2eb0635899410f538b734faf0e7a084a17b826c1acbd688f05ff", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.827589750289917, "target_z": 0.12695310651720573, "ce": 5.9431867599487305, "genuineness": 4.174716949462891, "identity": 0.06471633911132812, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/79d0edb4fdbff76125156485.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2e32478211cbd68b9403c39340e09f4b24e857f3c6a0fa55cbf0a4497b4a2d45", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.048753499984741, "target_z": 0.3007812379221344, "ce": 5.974957466125488, "genuineness": 4.143336772918701, "identity": 0.7048467993736267, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/14957dd4b95cb064750efff6.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "f89eb1b79d4809f7f2a826923c83722b3e41bd539c26bd6a5fb2dcaf1a21d4a3", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2475523948669434, "target_z": 0.4570312584179063, "ce": 5.911416053771973, "genuineness": 4.268857479095459, "identity": 0.6817120313644409, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/53c92e5ebf2a32b1577aa4b0.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "c62dc3eb9b1d699f0da51b7d3928518d9504d3a93ccdd91853b6547d35a29ae3", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.5623176097869873, "target_z": -0.08154290858941961, "ce": 6.157639026641846, "genuineness": 4.3504462242126465, "identity": 0.7749676704406738, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/cd2707fe42faa8eceb347756.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "2f2320eecc9eb072bef7bb07ccbf03d85c6e8ef66c29c2db0995ac1edfbb00f7", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2873120307922363, "target_z": 0.4882811500831121, "ce": 6.125868320465088, "genuineness": 4.519899368286133, "identity": 0.43400847911834717, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/9742c40f8570aa84f9a87c9e.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "95047df846edacbe06898235eb521cccc78b94494d45704a58c98ad5af2704a4", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.823862314224243, "target_z": 0.12402345259733197, "ce": 6.1894097328186035, "genuineness": 4.281409740447998, "identity": 0.7508370280265808, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/a500fc33f601440d7bc3d267.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "d2a492ee2999aa2670b6438360dc801c251c087c65f475862d908152136550d5", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.879774570465088, "target_z": 0.1679688235651812, "ce": 6.0305562019348145, "genuineness": 4.369274139404297, "identity": 0.7259483933448792, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/f1c9b1ba0ba4d765d72f98b1.mp3" }, { "key": "original-vn_S_MONO_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "305034e92917802eeb05695e3c22d8546dcf477acd9ba677b59b4fa402b79577", "task_id": "vn_S_MONO_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 2.9791738986968994, "target_z": 0.24609374011811, "ce": 6.062326908111572, "genuineness": 4.570107460021973, "identity": 0.579279899597168, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/1a87eb2b117408947b4e07cf.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "b8f12f1c4dd93925d42801c9ef3b3f37058be21a0b043dea26f8e5bca1177921", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2475523948669434, "target_z": 0.4570312584179063, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.8425017595291138, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/4bc20ba25d7a1b8b57c75b33.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "c6109413519ce3678e75b453585ac04da44a959a3804152983c780e752df75e0", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.5159308910369873, "target_z": 0.6679687767177026, "ce": 6.022613525390625, "genuineness": 4.193545341491699, "identity": 0.6587111949920654, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/1f49ba95937f8badaa03bfc7.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "a0004254c13494b3d86436e333dd2160128bcaec784c0ccac8a073d5b56f2a05", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.1779727935791016, "target_z": 0.4023437606138819, "ce": 6.181467056274414, "genuineness": 4.281409740447998, "identity": 0.6556845903396606, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/7b8cdbbc93576d27d72ce4b9.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "c70f055840a620e2d3f80992aaaff5e8b858512faa395f3707592e79769de59f", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.1680328845977783, "target_z": 0.3945312876975804, "ce": 6.1020402908325195, "genuineness": 4.2374773025512695, "identity": 0.7198721170425415, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/a483b674c3bc7cca14ac5cb8.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "0149398cf056cba75be61357d557ba51d569cdc3464459e31482c9ed5d5d5f57", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.381741523742676, "target_z": 0.5624999238728473, "ce": 6.260893821716309, "genuineness": 4.425758361816406, "identity": 0.5257614254951477, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/fad111afd247d474e05708d1.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "982ef020582b54e1933374cc0ea90651d62820004a7ffdedb0e143f4c4ec2d2d", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.4960508346557617, "target_z": 0.6523436434951855, "ce": 6.1020402908325195, "genuineness": 4.325342178344727, "identity": 0.7182972431182861, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/6ab1e6aa16ef117dd04d5fdc.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "da4d23ffeea10c99480196fe7a2320b6d1abf7da52bbfca7033cfd50a54b7923", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.2227025032043457, "target_z": 0.4374999824321955, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.45658543705940247, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_MONO_reg/nl/46b90fac994ff5021abb90d7.mp3" }, { "key": "original-vn_S_MONO_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_MONO_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": [ "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "fbbfcf86f25b66e918e3cea38a3e3e2e608c0b65b30a2f73e68bb29f78e68f9a", "mp3_sha256": "70ea33d825ed320e45fa752adbd18b5e536d64b32724feb7ccc0328a7291fc7b", "task_id": "vn_S_MONO_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong introspective monologue delivery, not monotone speech. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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(Strong introspective monologue delivery, not monotone speech. 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": 3.471201181411743, "target_z": 0.6328125548993889, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.6463906764984131, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_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_S_MONO_reg/nl/fc67f8a8a26fc39e25935428.mp3" } ], "source_standard_audio_unchanged": true }